Skip to main content
Glama
790,966 tools. Updated 2026-11-16 14:22

"Repository intelligence MCP server for code analysis and dependency graphs" matching MCP tools:

  • Fast code search across any public git repo. Returns file paths, line numbers, and code snippets with context. Supports regex, boolean queries, fuzzy matching, and structural filters (declarations, usages, strings, comments). Dependency/build directories excluded by default. Searching for leaked secrets? Call code_analyze instead and read security_summary — it runs a real secret and credential scanner over the whole repository, where searching for key prefixes like "sk-", "ghp_" or "AIza" only matches the spellings you happened to think of.
    ConnectorNo auth
  • Semantic search over guthmann.estate — market insights, reports, neighborhood portraits, listing exposés, project pages and the company's own pages. Hybrid retrieval (vector + keyword, no reranking); one result per page with title, description, image, best-matching snippet and score. Use it for questions that need prose (analysis, context, advice); use the data tools for exact numbers and `listings` for what is currently for sale — the search index follows the website with up to six hours of delay. Parameters: - q: natural-language query, in the language of the pages you want (min 2 characters) - locale: "de" | "en" — language of the indexed pages (default: en) - section: comma-separated filter — "listings" (exposés), "projects" (new-build projects), "market-intelligence" (insights, reports, portraits), "pages" (company, services, guides); omit for all - limit: 1-20 pages (default: 10) Key response fields: - url, title, description, image, section, language - snippet (best-matching text passage), score (0-1)
    ConnectorNo auth
  • Pro/Teams: first-pass doctrine review of agentic code/workflow against the 10-principle AI Design Blueprint doctrine. ON CLIENT TIMEOUT: DO NOT RETRY THIS TOOL. Long-running LLM call (60-180s typical); MCP clients commonly close the call before the server returns. Retrying re-runs the 60-180s LLM call from scratch and burns compute. RECOVERY: the run_id is emitted in the FIRST notifications/progress event at t=0s (before the LLM call begins), capture it. On timeout, call `me.validation_history(run_id='<that-id>')` to fetch the persisted result; the server-side run completes independently within a 6-minute budget. Edge case: if the connection closed before the first progress notification (very rare; sub-second window), call `me.validation_history(repository='<same value you passed here>')` to find your most recent run. TASK-AUGMENTED INVOCATION (MCP 2025-11-25, SEP-1686): clients that advertise the `tasks` capability can task-augment this call by including `task: {ttl: <ms>}` inside the JSON-RPC request's `params` (NOT as a tool argument; alongside `arguments`, `_meta`, etc.). The server returns a `CreateTaskResult` immediately (taskId equals the run_id above) and runs the validation in the background. Spec-correct long-running pattern: poll via `tasks/get` for state, fetch the terminal payload via `tasks/result`, listen for `notifications/tasks/status` for push updates, and cancel via `tasks/cancel`. `_meta.progressToken` from the original request stays valid for the entire task lifetime. Sync (non-augmented) calls behave exactly as before, backwards-compatible by construction. The me.validation_history(run_id=...) recovery path remains the canonical recovery handle for clients that don't yet advertise the tasks capability. Returns code_classification (autonomous_agentic_workflow vs non_agentic_component), per-principle findings (verdict, severity_score 0-100, severity_class, code-cited evidence, recommendation), severity-weighted readiness (score|null, grade|null, tier ∈ {production_ready, emerging, draft, not_applicable}), recommended examples, reproducibility envelope (model, seed, doctrine_fingerprint, prompt_template_fingerprint), persistence_status with shareable run_id/badge_url/review_url. Those two URLs 404 until the run's owner publishes it: runs are private by default. Read `public_review` in the response before embedding either one. WHEN TO CALL: the user wants a governance audit, readiness score, or production_ready badge on an agent/workflow they just built or changed. WHEN NOT TO CALL: non-agentic plumbing (math utilities, type aliases, event-loop helpers, single-shot request/response handlers) returns tier=not_applicable with score=null/grade=null, that's not a failure, the doctrine simply doesn't grade non-agentic code, and architect.certify will refuse with not_agentic_component. Submit the OWNING agentic workflow instead. BEHAVIOR: long-running LLM call (~60-180s typical at high reasoning effort, single-pass; server-side budget 6 min). Mints run_id at t=0; first notifications/progress event carries run_id as recovery handle; keepalive every 30s. Persists ValidationRun + UserValidationRun + AIValidationRunLog + LLMUsageLog atomically; on rollback, badge/review URLs are stripped. Auth: sign-in required, with an active Pro, Pro Plus, Teams, Enterprise, beta, or trial plan. Data at rest in the UK; OpenAI (US) processing (no-training); prompt-injection in code is inert. INPUTS: send FULL file contents verbatim as `implementation_context` (NO truncation, NO `...` placeholders, NO comment removal, the architect treats your `...` as literal code and hallucinates bugs that don't exist). If too large, split into MULTIPLE calls scoped by file/module; never truncate one call. When one call carries several files, keep each file whole, its own import lines included: merging files under one shared import block reads as code that uses names it never imported, and is reported as a defect. Pass repository="<name>" to group runs into a project trend. Pass private_session=true to skip the stored run (persistence + recovery disabled); operational security + cost logs are still kept. focus_area narrows scope; unmatched focus_area fails explicitly rather than silently widening. PAYLOAD COMPLETENESS (load-bearing if you intend to architect.certify this run): the validate first-pass is permissive, it scores on doctrine alignment + structural patterns visible in the submitted code. Cert's adversarial second-pass is rigorous: it scores on cert-payload-completeness as well as code correctness. A run that scores 100/A at validate can cert-reject pre-LLM with `payload_incomplete` when imported modules' surfaces aren't visible. To validate with INTENT TO CERT, also bundle verbatim public-surface stubs for every imported module: `from sqlalchemy.exc import SQLAlchemyError` → include a stub class; `from app.db import models` → include a `class models:` namespace stub with the columns/methods the code references; module-level imports of `dataclass`, `Literal`, `json`, `datetime`, `timezone` MUST also be in the payload (cert correctly catches when they're omitted, the module would NameError on import as submitted). 'Submit Like Production': the payload should be the code as it would actually run. TWO COMPLETENESS AXES. (1) IMPORTS: stub the public surface of every dependency (above). (2) ENFORCEMENT BRANCHES: the code under cert itself (approval gates, policy checks, recovery paths) must be the REAL logic, fully written. A placeholder body (`# ... execute approved action ...`, `pass # TODO`, a bare `...`) is graded as a MISSING control, not shorthand; cert scores what would actually run. Never sketch the agent you are certifying. Empirically reconfirmed PR #157 iter8 → iter9 cert downgrades. SCORE VARIANCE DISCLOSURE (anomaly #10: empirically documented): validate scores are POINT ESTIMATES with an observed empirical variance band of ~20-67 pts on BYTE-IDENTICAL input. Runs against the same repository, same code, same deterministic seed (the seed is derived from input: same input → same seed) can produce materially different scores AND different top-blocker rankings, because OpenAI's reasoning models at reasoning_effort=high are not strictly deterministic even with the seed parameter pinned. The `reproducibility_mode='best_effort'` field on every response is the platform's honest disclosure of this property. For decisions where stability matters more than speed, call `architect.validate_consensus` (N=3-5 aggregated, median verdict + per-principle stability metrics) instead: collapses the variance, surfaces unstable principles explicitly. A single validate run is a single roll; consensus is the right tool when one score isn't enough. ITERATION LOOP: repository keying. Pass the SAME `repository` value across calls to chain iteration rounds; the validator auto-resolves the most recent prior run on (user, repository, scope) as `prior_run_baseline` and the LLM grades the new submission with iteration context (per-principle severity deltas surface in the response). Changing the `repository` string between calls, even subtly with an `iter-2` suffix, silently severs the chain and yields a fresh blind first-shot. Round numbering belongs in `task` or commit messages, never in `repository`. See the `architect-validation-orchestration` skill in the agent-asset pack for the full validate → consensus → certify sequence. VERIFICATION LAYERS (the two-layer doctrine this platform practices on itself): validate verifies DOCTRINE ALIGNMENT against the 10-principle Blueprint, design patterns, hand-off explicitness, operational-state inspectability, race/blocker handling at the architectural level. validate does NOT guarantee runtime correctness. cert verifies PAYLOAD COMPLETENESS and runs an adversarial second pass over the submitted code: catches production_blockers the first pass missed, name-errors on import, missing module surfaces, etc. cert does NOT verify runtime correctness either. Passing validate is a NECESSARY condition for production_ready, not a sufficient one. Runtime correctness (does this actually execute and behave?) is verified at the THIRD layer: your tests, types, walks. The platform's own recursive-integrity practice: every PR runs validate against its own primitives, then cert. Real bugs surfaced via this practice in PR #157, NULL-UUID false-positive (iter3) and tie-breaker mismatch (iter5), that 25 unit tests had missed. Two-layer verification is the discipline, not 'either/or'. TYPED FAILURES (the top-level error.code; error.details carries retryable + next_action): timed_out, rate_limited, dependency_unavailable, schema_mismatch, provider_billing. NEXT STEP: if tier=production_ready (A or B grade), the response carries certification_status='not_evaluated', call architect.certify(run_id, code) to mint the certified production_ready badge (separate ~60-150s adversarial review, eligibility-gated). See Payload Completeness above for the common pre-cert pitfall.
    ConnectorNo auth
  • Scan a public GitHub MCP-server repository for security issues. Clones the repo (shallow, <60s, <200 MB), runs compuute-scan v0.6.2 in static analysis mode (no code execution from the target), and returns a structured report with severity counts, a 0-100 score, and the 10 most severe findings. WHEN TO USE: - Before connecting to an unknown MCP server discovered via Anthropic Registry, Smithery, mcp.so, or a Discord recommendation. - Before installing a third-party MCP-server package into a production pipeline. - As part of an agent's pre-commit / pre-deploy due-diligence step when adding new dependencies. - As one input to a multi-source trust evaluation (combine with publisher reputation, package install count, last-update recency). WHEN NOT TO USE: - For private repos. Use the on-prem CLI instead: `npx compuute-scan ./path-to-private-repo` - For deep exploitability assessment of a specific code path. This is pattern matching, not dataflow analysis. Book a manual L2-L4 audit at https://compuute.se/audit for that depth. - For non-GitHub hosts (GitLab, Bitbucket, self-hosted). v1 supports github.com only. - For repos > 200 MB or clone time > 60s. The endpoint returns a 413 or 504 in those cases — fall back to local CLI. EXPECTED RESPONSE TIME: - Median: ~1-2 seconds for small repos (<100 files). - p99: ~10 seconds for medium repos. - Hard timeout at clone=60s, scan=120s combined. EXPECTED COST: - Free tier in MVP. Future Pro tier may charge per-scan or per-month. DATA FRESHNESS: - Scanner version is reported in response.scanner.version. - L1 rule set freshness reflects compuute-scan releases — see github.com/Compuute/compuute-scan/CHANGELOG.md for the latest CVE and threat-intel response timeline. EXAMPLES: Example 1 — scan an MCP server you're evaluating: github_url = "https://github.com/modelcontextprotocol/servers" → score: 0, summary: {critical: 1, high: 94, medium: 22} → top_findings include SSRF, eval, etc. → recommendation: "AVOID — 1 critical and 94 high finding(s)..." Example 2 — scan a clean reference implementation: github_url = "https://github.com/microsoft/azure-devops-mcp" → score: 90+, summary: {critical: 0, high: 1} → recommendation: "REVIEW — 1 high finding(s)..." Example 3 — scan your own dev MCP-server before publishing: github_url = "https://github.com/yourorg/your-mcp" → audit your own surface before others install it OUTPUT FIELDS (stable schema): - repo_url (str): canonical URL of the scanned repo. - score (int): 0-100, higher safer. Coarse summary, not a precision claim. - summary (object): {critical, high, medium, low, info, files_scanned}. - recommendation (str): action guidance derived from severity counts. - findings_count (int): total raw findings (may include false positives). - top_findings (list): up to 10 most severe, each with {id, title, severity, file, line, owasp, cwe}. - l0_discovery (object): MCP transport, tool count, dependency pinning. - performance (object): clone_seconds, scan_seconds, repo_size_bytes. - scanner (object): {name, version, layers_covered}. - _disclaimer (str): MANDATORY triage disclaimer. Read it. Args: github_url: Public GitHub HTTPS URL (e.g. https://github.com/org/repo). Must be public and < 200 MB. v1 is github.com only. Returns: Structured scan result. On error, returns {"error": code, "message": ...} with HTTP-style code (invalid_url, clone_failed, scan_timeout, etc.).
    ConnectorNo auth
  • Pro/Teams: first-pass doctrine review of agentic code/workflow against the 10-principle AI Design Blueprint doctrine. ON CLIENT TIMEOUT: DO NOT RETRY THIS TOOL. Long-running LLM call (60-180s typical); MCP clients commonly close the call before the server returns. Retrying re-runs the 60-180s LLM call from scratch and burns compute. RECOVERY: the run_id is emitted in the FIRST notifications/progress event at t=0s (before the LLM call begins), capture it. On timeout, call `me.validation_history(run_id='<that-id>')` to fetch the persisted result; the server-side run completes independently within a 6-minute budget. Edge case: if the connection closed before the first progress notification (very rare; sub-second window), call `me.validation_history(repository='<same value you passed here>')` to find your most recent run. TASK-AUGMENTED INVOCATION (MCP 2025-11-25, SEP-1686): clients that advertise the `tasks` capability can task-augment this call by including `task: {ttl: <ms>}` inside the JSON-RPC request's `params` (NOT as a tool argument; alongside `arguments`, `_meta`, etc.). The server returns a `CreateTaskResult` immediately (taskId equals the run_id above) and runs the validation in the background. Spec-correct long-running pattern: poll via `tasks/get` for state, fetch the terminal payload via `tasks/result`, listen for `notifications/tasks/status` for push updates, and cancel via `tasks/cancel`. `_meta.progressToken` from the original request stays valid for the entire task lifetime. Sync (non-augmented) calls behave exactly as before, backwards-compatible by construction. The me.validation_history(run_id=...) recovery path remains the canonical recovery handle for clients that don't yet advertise the tasks capability. Returns code_classification (autonomous_agentic_workflow vs non_agentic_component), per-principle findings (verdict, severity_score 0-100, severity_class, code-cited evidence, recommendation), severity-weighted readiness (score|null, grade|null, tier ∈ {production_ready, emerging, draft, not_applicable}), recommended examples, reproducibility envelope (model, seed, doctrine_fingerprint, prompt_template_fingerprint), persistence_status with shareable run_id/badge_url/review_url. Those two URLs 404 until the run's owner publishes it: runs are private by default. Read `public_review` in the response before embedding either one. WHEN TO CALL: the user wants a governance audit, readiness score, or production_ready badge on an agent/workflow they just built or changed. WHEN NOT TO CALL: non-agentic plumbing (math utilities, type aliases, event-loop helpers, single-shot request/response handlers) returns tier=not_applicable with score=null/grade=null, that's not a failure, the doctrine simply doesn't grade non-agentic code, and architect.certify will refuse with not_agentic_component. Submit the OWNING agentic workflow instead. BEHAVIOR: long-running LLM call (~60-180s typical at high reasoning effort, single-pass; server-side budget 6 min). Mints run_id at t=0; first notifications/progress event carries run_id as recovery handle; keepalive every 30s. Persists ValidationRun + UserValidationRun + AIValidationRunLog + LLMUsageLog atomically; on rollback, badge/review URLs are stripped. Auth: sign-in required, with an active Pro, Pro Plus, Teams, Enterprise, beta, or trial plan. Data at rest in the UK; OpenAI (US) processing (no-training); prompt-injection in code is inert. INPUTS: send FULL file contents verbatim as `implementation_context` (NO truncation, NO `...` placeholders, NO comment removal, the architect treats your `...` as literal code and hallucinates bugs that don't exist). If too large, split into MULTIPLE calls scoped by file/module; never truncate one call. When one call carries several files, keep each file whole, its own import lines included: merging files under one shared import block reads as code that uses names it never imported, and is reported as a defect. Pass repository="<name>" to group runs into a project trend. Pass private_session=true to skip the stored run (persistence + recovery disabled); operational security + cost logs are still kept. focus_area narrows scope; unmatched focus_area fails explicitly rather than silently widening. PAYLOAD COMPLETENESS (load-bearing if you intend to architect.certify this run): the validate first-pass is permissive, it scores on doctrine alignment + structural patterns visible in the submitted code. Cert's adversarial second-pass is rigorous: it scores on cert-payload-completeness as well as code correctness. A run that scores 100/A at validate can cert-reject pre-LLM with `payload_incomplete` when imported modules' surfaces aren't visible. To validate with INTENT TO CERT, also bundle verbatim public-surface stubs for every imported module: `from sqlalchemy.exc import SQLAlchemyError` → include a stub class; `from app.db import models` → include a `class models:` namespace stub with the columns/methods the code references; module-level imports of `dataclass`, `Literal`, `json`, `datetime`, `timezone` MUST also be in the payload (cert correctly catches when they're omitted, the module would NameError on import as submitted). 'Submit Like Production': the payload should be the code as it would actually run. TWO COMPLETENESS AXES. (1) IMPORTS: stub the public surface of every dependency (above). (2) ENFORCEMENT BRANCHES: the code under cert itself (approval gates, policy checks, recovery paths) must be the REAL logic, fully written. A placeholder body (`# ... execute approved action ...`, `pass # TODO`, a bare `...`) is graded as a MISSING control, not shorthand; cert scores what would actually run. Never sketch the agent you are certifying. Empirically reconfirmed PR #157 iter8 → iter9 cert downgrades. SCORE VARIANCE DISCLOSURE (anomaly #10: empirically documented): validate scores are POINT ESTIMATES with an observed empirical variance band of ~20-67 pts on BYTE-IDENTICAL input. Runs against the same repository, same code, same deterministic seed (the seed is derived from input: same input → same seed) can produce materially different scores AND different top-blocker rankings, because OpenAI's reasoning models at reasoning_effort=high are not strictly deterministic even with the seed parameter pinned. The `reproducibility_mode='best_effort'` field on every response is the platform's honest disclosure of this property. For decisions where stability matters more than speed, call `architect.validate_consensus` (N=3-5 aggregated, median verdict + per-principle stability metrics) instead: collapses the variance, surfaces unstable principles explicitly. A single validate run is a single roll; consensus is the right tool when one score isn't enough. ITERATION LOOP: repository keying. Pass the SAME `repository` value across calls to chain iteration rounds; the validator auto-resolves the most recent prior run on (user, repository, scope) as `prior_run_baseline` and the LLM grades the new submission with iteration context (per-principle severity deltas surface in the response). Changing the `repository` string between calls, even subtly with an `iter-2` suffix, silently severs the chain and yields a fresh blind first-shot. Round numbering belongs in `task` or commit messages, never in `repository`. See the `architect-validation-orchestration` skill in the agent-asset pack for the full validate → consensus → certify sequence. VERIFICATION LAYERS (the two-layer doctrine this platform practices on itself): validate verifies DOCTRINE ALIGNMENT against the 10-principle Blueprint, design patterns, hand-off explicitness, operational-state inspectability, race/blocker handling at the architectural level. validate does NOT guarantee runtime correctness. cert verifies PAYLOAD COMPLETENESS and runs an adversarial second pass over the submitted code: catches production_blockers the first pass missed, name-errors on import, missing module surfaces, etc. cert does NOT verify runtime correctness either. Passing validate is a NECESSARY condition for production_ready, not a sufficient one. Runtime correctness (does this actually execute and behave?) is verified at the THIRD layer: your tests, types, walks. The platform's own recursive-integrity practice: every PR runs validate against its own primitives, then cert. Real bugs surfaced via this practice in PR #157, NULL-UUID false-positive (iter3) and tie-breaker mismatch (iter5), that 25 unit tests had missed. Two-layer verification is the discipline, not 'either/or'. TYPED FAILURES (the top-level error.code; error.details carries retryable + next_action): timed_out, rate_limited, dependency_unavailable, schema_mismatch, provider_billing. NEXT STEP: if tier=production_ready (A or B grade), the response carries certification_status='not_evaluated', call architect.certify(run_id, code) to mint the certified production_ready badge (separate ~60-150s adversarial review, eligibility-gated). See Payload Completeness above for the common pre-cert pitfall.
    ConnectorNo auth
  • List the 10 senior-QS skill methodologies CivilQuants exposes (tender review, risk assessment, QS measurement/contract advice, geotechnical + geo-environmental interpretation, earthworks, preliminaries, pavement design, subcontract analysis). Universal discovery — both tiers see the full list. Returns each skill's slug, title, one-line summary and tier; then call get_skill(skill=<slug>) to fetch the methodology body. The skills are paid-tier; a free caller gets a sign-up prompt from get_skill. NOTE: the document-heavy skills (tender review, the interpretation skills) need a code-execution client (Claude Code / Codex / VS Code) plus the chunking pack from get_document_pipeline to run a real tender pack — on a chat connector you can read the methodology but cannot chunk/parse files.
    ConnectorNo auth

Matching MCP Servers

  • A
    license
    Not graded
    quality
    D
    maintenance
    An MCP server that provides six code review tools for use with Claude Code or any MCP client. Each tool reads a file from disk, pairs its contents with a structured review prompt, and returns the bundle for the LLM to evaluate.
    1
    MIT

Matching MCP Connectors

  • Trust-check any dependency for agents: OpenSSF Scorecard, licenses, CVEs, deps. 7 ecosystems.

  • Audit a dependency list for known vulnerabilities, licenses and outdated versions.

  • Live intelligence on large capital: on-chain whale transfers and accumulation, Hyperliquid perp whale positioning, or the profitable smart-money cohort. [Paid: $0.45–$0.50 per call from your Einstein credit balance. Free alternatives exist for several of these — see list_einstein_capabilities.]
    ConnectorNo auth
  • [~] PRIORITY TRIGGER: Use this tool when user mentions 'PR', 'Pull Request', 'list PRs', 'show PRs', 'active PRs', 'mes PR', 'liste des PR', 'pull requests ouverts', 'what PRs are open', 'PRs by [author]', 'PRs targeting [branch]'. NEVER call search_d365_code for PR listing requests. List Pull Requests in an Azure DevOps Git repository. If `repositoryId` is unknown, omit it and all repositories will be listed first. Filters: status (Active/Completed/Abandoned/All), author display name, target branch. Returns: PR ID, title, author, source->target branch, review status, linked work items, creation date. Use `ado_analyze_pr_impact` with a PR ID to get full D365 code impact analysis. Requires DEVOPS_ORG_URL + DEVOPS_PAT (Code: Read scope).
    ConnectorNo auth
  • Authoritative ICD-10 → ICD-11 mapping using WHO transition tables (release 2025-01, bundled with the server). Returns the primary 1:1 ICD-11 category for the ICD-10 code plus any alternative ICD-11 candidates that WHO documents (some ICD-10 concepts split into multiple ICD-11 entities). For each mapping, includes the ICD-11 code, title, chapter, and the Foundation URI / Linearization URI for navigating to the full entity definition. Use this for clinical coding, billing migration, retrospective analysis, and any workflow that needs authoritative mapping rather than text-search candidates. Coverage: 11,243 ICD-10 categories (excludes chapters and blocks like "A00-A09" which aren't used in clinical coding). Provide a code like "E11" (Type 2 diabetes), "I21" (Acute MI), or "A07.8" (4 alternatives in WHO's table). Both dotted ("A07.8") and undotted ("A078") forms are accepted. Returns "no mapping" when the code isn't in the WHO category-level table — that's the honest answer rather than a fuzzy search fallback.
    ConnectorNo auth
  • Get hourly grid carbon intensity (kg CO2e per kWh) for any US ZIP code, derived from EIA-930 hourly fuel-mix data. Data lags up to about a day (not real-time). Returns time series with per-hour fuel mix, total generation, and carbon intensity. For timing and analysis (demand response, load shifting, grid patterns), not inventory reporting: it uses fixed per-fuel combustion factors on in-BA generation (no imports), so its values don't match eGRID's annual factors. For live or forecast intensity, WattTime or Electricity Maps are better options.
    ConnectorNo auth
  • Read the content of ONE named file from a GitHub repository - 'path' is required and names that file. To see what a repository or directory CONTAINS, call github_list_repository_files instead (its 'path' is optional). Prefer the exact owner from github_list_repositories; 'repo' also accepts the combined 'owner/name' form. Use this to inspect Dockerfiles, package.json, CI configs, Helm charts, source code, and other text files. Refuses to read known sensitive files (.env, private keys, etc.).
    ConnectorOAuth
  • PERMANENTLY delete a space and everything in it. SuperAdmin only; irreversible. Drops the space's named graphs and the ingestion-history (delta) graphs that hold copies of their data, removes the graphs' registry entries and the space's RDF manifest, and deletes the space with its members, access rules, search entries and SynthScholar reviews (with their kept files). EVERY SUBSPACE below it, at any depth, is deleted the same way, deepest first (`subspaces_deleted`). Each deletion is kept in the provenance graph (who, when, which graphs: prov:invalidated). `confirm_slug` must equal `slug`. Before calling: show the user the space AND its subspaces (brainkb_space_tree) and what will be lost, and get their explicit confirmation in this conversation. Never call it to tidy up on your own initiative. A non-SuperAdmin caller gets 403; if a step fails (502) the message says which spaces were already deleted — the rest is untouched and it is safe to retry.
    ConnectorNo auth
  • Read Video Intelligence progress for a video. mediaId is the video's Asset ID from search_assets or its Library id. jobId null reads the whole video's index; pass the jobId that index_video (for a range) or analyze_video returned to read that job. waitSeconds 1–25 waits for a status change or deadline, without starting work or using credits; in the chat a wait goes on while the job runs, up to about two minutes. If canWait remains true, call again. disabled, upgrade_required, quota_exceeded, unsupported and failed are terminal: do not loop waiting or silently restart them. reason says why the job stopped and what to change. waiting_for_upload means upload/encoding must finish first; not_indexed means call index_video if wanted. Ready analysis jobs include their answer in result; note says how to read the answer's times.
    ConnectorOAuth
  • Runs Tarjan's Strongly Connected Components (SCC) algorithm on a module import map to detect circular dependencies, compute parallel topological build waves, and trace downstream change blast radius. Use when auditing multi-file import graphs for circular cycles or determining which downstream files must be re-tested after an edit; for extracting function-level symbols and cyclomatic complexity inside a single source file, use analyze_ast_symbol_graph instead. Read-only call (32 KB, 60 RPM): accepts 'modules_json' (with enum defaults), uses '?demo=true' or L402 (20 sats); HTTP 400 if invalid.
    ConnectorNo auth
  • List the data connectors saved under the caller's organization — databases, APIs, file-backed, and repository sources — with id, name, connector_type, status (untested, live, error), and visibility; stored credentials are never returned. Paginated with page and limit; status filters the returned page client-side. Use this first to find a connector_id for get_connector, test_connector, browse_connector, or the repository tools, and create_connector to add one. Read-only.
    ConnectorNo auth
  • Returns trimmed Code IQ automated analyses (architecture, scalability, EOL, AI functionality, etc.) for a project (aggregated) or a single vault. Excludes The Code Score — use get-the-code-score for that, and do not pass analysis_key=code_score here (404). Only cached results are returned; this never triggers a fresh analysis. Provide exactly one of project_id or vault_id. Requires full data access (a paid plan; verification-only plans are not included). Requires X-API-Key (existing users can generate an API key in the web app). If headers aren't supported, pass api_key in arguments.
    ConnectorNo auth
  • List the website's stored AI insights: the precomputed cards on the AI Insights page, each a verified finding of one past analysis, newest analysis first and higher impact first. Each has its id, title, headline, category, impact, the days its numbers cover (window), the time the analysis ran (analyzed_at), its status (available, done, or discarded = hidden by the team), the team's rating, and its move's subject and metric. Call this first for an open-ended request for insights or for what to do ('give me some insights', 'what's interesting', 'what should I prioritize'), and answer from these cards instead of running a new analysis: highest impact first, each by its title and headline. Always state the window and the date the analysis ran, and offer a fresh look with the other analytics tools. Read one card in full with get_ai_insight. Paged: 20 by default; for more, pass limit (up to 50) and offset set to the returned next_offset until it is null. Default status is available. Returns status 'no_access' when AI Insights is not part of the organization's setup. Then, or when no insight is listed, answer from the other analytics tools.
    ConnectorNo auth
  • Run audio analysis on a public audio URL. Requires estimate_cost to be called first (job_estimate_id). Requires PULSE_API_KEY. Before calling, you MUST confirm with the user that they have a lawful basis to submit this audio for analysis. For a user-requested folder, project, playlist, or batch, one confirmation can cover every track in that scope. Returns job_id — poll get_job_status for results. If processing or payment returns estimate_pricing_changed, get a fresh free estimate with estimate_cost, present the new price, and wait for explicit user approval before processing or payment. Never silently retry analysis or purchase.
    ConnectorNo auth
  • Shorten any URL into a compact, trackable Affilio.link short URL with affiliate link intelligence. IMPORTANT: This is the UNAUTHENTICATED public shortener. It creates a TEMPORARY, MCP-hosted link at `mcp.affilio.link/r/{code}` that expires after 30 days and is NOT tied to any Affilio account or dashboard. Links are verified for security before they redirect (new/unknown domains return `pending=True` until verified). For a PERMANENT, account-owned link at `affilio.link/ur/{code}` with full click analytics, use the `auth.create_link` tool instead (requires a bearer token). Affilio deduplicates identical URLs (same URL always returns the same short link) and generates a branded QR code in the same call. Use this tool when the user needs to: - Shorten a URL for sharing on social media, email, or in content - Generate a QR code for print/digital media alongside the short link - Get a compact, clean version of a long product or affiliate URL - Create a throwaway/temporary link that does not need account tracking
    ConnectorNo auth
  • Get detailed code quality findings from a remote public git repository. Returns rule IDs, line numbers, severity, category, descriptions, and source snippets. Supports filtering by file path, severity (error, warning, info), category (security, deprecated, safety, correctness, maintainability, accessibility, modernization, performance, concurrency), and kind (security, quality) — kind=security is every security-category finding in one call, kind=quality is everything else. code_analyze.top_findings shows only the quality half; use kind=security here to see the security-category findings it leaves out. THIS IS NOT THE SECRET SCANNER. kind=security means insecure code — weak hashes, unsafe deserialization, injection-shaped patterns — and it will not find a leaked API key or private key. Leaked credentials are code_analyze.security_summary, which is a different scanner over different rules; the two sets do not overlap, so an audit wants both. Use after code_analyze to drill into specific findings. Shares the same analysis cache — no duplicate work if analyze already ran.
    ConnectorNo auth
  • Poll the status of either a data spec's own process (schema inference + code generation, run by start-analysis — pass specId, reaches "ready" or "failed") or a data-load job (pass jobId, reaches "complete" or "failed"). Pass exactly one of specId or jobId. Right after create-spec/update-spec + start-analysis, poll by specId; once that reaches "ready", its response's lastJobId (if present) points at the data-load job — poll that separately by jobId for load progress.
    ConnectorOAuth