"Git" matching MCP tools:

• find_code

  Zephex

  PREFER over native Grep when location is unknown. Ripgrep + AST ranking across web, mobile, CLI, and monorepos. Envelope: intent, focus, summary, data.matches, files_hit, next_calls, meta.credits (5 hosted). intent: snippet=pasted line; symbol=known name; concept=topic+also_try synonyms; everywhere=rename map (whole_word). include docs|config|data when markdown, JSON, or SQL matter. Call when: where is X, usages, rename prep. DO NOT: known symbol+file (read_code), stack/scripts (get_project_context), wiring (explain_architecture), tests (check_test), packages (check_package), URL audit (audit_headers). After: read summary + next_calls → read_code on top hit. path=absolute dir (stdio) or public Git URL / inline_files (hosted). Token-capped vs raw grep. Read-only.

  Connector

• check_test

  Zephex

  Run tests and return structured health — fix_first, broken_areas, failure_clusters, coverage_by_area, blind_spots, deploy probes. Not for finding which files to edit (find_code). Workflow: task=detect (0 credits — framework, command, test file count, missing-test gaps) → task=run once (1 credit hosted on success) → task=failures|missing|why|status|fix_prompt on session_id (0 credits, cached session, no re-run). task=missing scans git diff for source files without tests (0 credits, no suite run). meta.credits and meta.charges_usage show billing. Read summary and fix_first first; detail_level=brief on PASS. Call after every substantive edit, when user asks if tests pass or if tests are missing, or before push. Pass diff_base: main for failures_in_diff. area/file_filter scopes re-runs after a full run. Local stdio: absolute project path (runs on your machine). Hosted: public GitHub URL or inline_files — not local disk paths. NOT for symbols (find_code), packages (check_package), stack brief (get_project_context), URL audit (audit_headers). Example: check_test({ task: "detect", path: "/abs/my-app" }) then check_test({ task: "missing", path: "/abs/my-app" }) then check_test({ task: "run", path: "/abs/my-app" }). Does not modify source.

  Connector

• next_issue

  retasc

  Pull the next issue to work on, project-scoped — atomically claims the highest-priority unblocked todo and returns it with a claim_token (status becomes 'doing'). Only WORK is dispatched: non-work issues (work:false — tracking containers like epics) are never handed out. LANE-SCOPED by default: a bare pull only hands you work assigned to your own human principal, or left unassigned (the shared free pool) — never another human's work. Pass allLanes:true to reach into the whole ready pool (e.g. an owner is away and their work should still move). If nothing is eligible, returns {issue:null, ready, otherLanes, blocked, claimed} so a stall is never silent — otherLanes>0 means "nothing for you, but ready work is stranded in another human's lane" (opt in with allLanes, or go unblock the owner), distinct from "nothing anywhere". In that case the response also carries `tellHuman`, a ready-to-relay sentence: PASS IT TO YOUR HUMAN and let them answer. Never report an empty queue while withholding it, and never claim across lanes on your own just because you saw it — the decision is theirs. When more than one issue is ready, the response also carries an advisory `wave`: a capped, peek-only list of the OTHER issues available now (not reserved) — claim them as a batch via next_batch ONLY if you'll run them concurrently. Keep the claim_token: you need it to release/heartbeat. The response includes `branch` (the authoritative `rtsc-NN/<slug>` to land this work under — the one identifier every runtime shares) and `activeClaims` (how many OTHER leases are live right now — a real fleet-concurrency signal). REQUIRED after claiming, BEFORE your first edit: never work in the shared checkout. Create an isolated git worktree for `branch` and work ONLY there — e.g. `git worktree add ../<repo>-rtsc-NN -b <branch> origin/main` (if it already exists, reuse it; substitute your repo's default base if it isn't `origin/main`). Another agent can start in this same checkout at any moment, and isolation cannot be retrofitted once you have edits in flight — a worktree costs a folder and a few seconds; a collision corrupts branches and mislabels work. Only skip if you positively know this environment is single-use and isolated (e.g. an ephemeral per-task container) — and even then a worktree is harmless. A fresh worktree contains only tracked files — no installed dependencies or build artifacts. Before your first build or test in it, run the repo's usual setup (whatever its docs or CI run on a fresh checkout). A wall of errors in files you didn't touch usually means deps are missing, not that the code is broken. Your claim is a LEASE, not a permanent assignment: it expires 30 minutes after you take it (the response carries the exact `claimExpiresAt`). ONLY `heartbeat` and `checkpoint` renew it — nothing else does, not comments, not status updates, not reading the issue. Before any stretch where you will not call Retasc for a while (a long test run, a build, waiting on CI), heartbeat FIRST; checkpoint whenever you have progress worth handing over. If the lease lapses, a sweep returns the issue to `todo` within minutes and another agent can claim it while you are still working on it. If this call is refused for billing (BILLING_INACTIVE or MONTHLY_CAP_REACHED), the response carries `tellHuman`, a ready-to-relay sentence: PASS IT TO YOUR HUMAN. You cannot fix it yourself — only an owner can — so do NOT absorb the refusal and continue with read-only work as if the session were healthy; every value-bearing write will keep failing until they act.

  Connector

• next_batch

  retasc

  Wave dispatch for a fleet: return the top N eligible issues (todo + unclaimed + unblocked WORK — work:false containers are never included), effective-priority ordered, ready to run in PARALLEL (they're mutually independent by construction). Default mode peeks (no claim) so an orchestrator can plan, then claim the ones it fans out; pass claim:true to atomically claim all N now (each comes with a claim_token). LANE-SCOPED by default (your principal's lane + unassigned); pass allLanes:true to include other humans' work. Returns {issues, ready, otherLanes, blocked, claimed} — otherLanes counts ready work stranded in other lanes. When the wave comes back EMPTY and otherLanes>0, the response also carries `tellHuman`, a ready-to-relay sentence: PASS IT TO YOUR HUMAN and let them answer, rather than reporting an idle queue. Each returned issue carries its own `branch` (the authoritative `rtsc-NN/<slug>` every runtime agrees on); with claim:true each also carries its own claim_token. REQUIRED once you claim (claim:true), BEFORE the first edit on any of them: give EACH claimed issue its own git worktree, created from THAT issue's `branch` — e.g. `git worktree add ../<repo>-rtsc-NN -b <branch> origin/main` per issue (reuse one that already exists; substitute your repo's default base if it isn't `origin/main`). Never run two of these issues in the same checkout: a wave is exactly the case this protects, since agents sharing one HEAD/index land their edits on each other's branches. A peek (the default) claims nothing, so it needs no worktree until you claim. Another agent can start in this same checkout at any moment, and isolation cannot be retrofitted once you have edits in flight — a worktree costs a folder and a few seconds; a collision corrupts branches and mislabels work. Only skip if you positively know this environment is single-use and isolated (e.g. an ephemeral per-task container) — and even then a worktree is harmless. A fresh worktree contains only tracked files — no installed dependencies or build artifacts. Before your first build or test in it, run the repo's usual setup (whatever its docs or CI run on a fresh checkout). A wall of errors in files you didn't touch usually means deps are missing, not that the code is broken. Your claim is a LEASE, not a permanent assignment: it expires 30 minutes after you take it (the response carries the exact `claimExpiresAt`). ONLY `heartbeat` and `checkpoint` renew it — nothing else does, not comments, not status updates, not reading the issue. Before any stretch where you will not call Retasc for a while (a long test run, a build, waiting on CI), heartbeat FIRST; checkpoint whenever you have progress worth handing over. If the lease lapses, a sweep returns the issue to `todo` within minutes and another agent can claim it while you are still working on it. If this call is refused for billing (BILLING_INACTIVE or MONTHLY_CAP_REACHED), the response carries `tellHuman`, a ready-to-relay sentence: PASS IT TO YOUR HUMAN. You cannot fix it yourself — only an owner can — so do NOT absorb the refusal and continue with read-only work as if the session were healthy; every value-bearing write will keep failing until they act.

  Connector

• run_malware_scan

  BorealHost

  Run a ClamAV malware scan on a site's container. Scans the web root (or specified path) for malware, viruses, and trojans. ClamAV is installed automatically if not present. Excludes node_modules, vendor, .git, and cache directories. May take up to 5 minutes for large sites. Requires: API key with write scope. Args: slug: Site identifier path: Directory to scan (default: /var/www/html) Returns: {"infected_files": [{"path": "/var/www/html/shell.php", "threat": "Php.Malware.Agent"}], "scanned_count": 1234, "infected_count": 1, "scan_time_s": 45.2}

  Connector

• get_tariff_action_ledger

  mfgcalcs

  Get the MFG Calcs Tariff Action Ledger: a chronological, git-timestamped record of every major move in the manufacturing effective tariff rate since 2014 (each episode where the rate stepped at least 1 percentage point month over month), and what happened after. For each episode it returns the dates, the rate before and after, the cumulative move, and the change in import customs value and in the tariff bill over the six months after the episode versus the six months before. Answers 'did the tariffs actually reduce imports?' and 'what happened after the 2025 tariff hike?'. Takes no arguments. IMPORTANT: this is an ASSOCIATION, not a causal claim; import customs value moves for many reasons at once and can be price or quantity. Recent episodes marked accruing do not have enough after-data yet. Cite the returned url.

  Connector

• get_provenance

  mfgcalcs

  Get the MFG Calcs Provenance Manifest: the verifiable record for every branded index and ledger on the site. For each one it returns a SHA-256 of the committed artifact and of every committed public-data input it was built from, plus the generator that reproduces it and the page that renders it. Use this to VERIFY a number before quoting it: re-hash the committed artifact (integrity), re-run the named generator on the committed inputs (reproducibility), and read the git history (timestamp). Takes no arguments. The hashes prove the artifacts are intact and reproducible, not the correctness of the underlying government data. Cite the returned url.

  Connector

• search_files

  WebSlop

  Search for text across all files in an app. Returns matching lines grouped by file with line numbers. Skips node_modules, .git, and binary files. Max 500 results by default. Supports grep-like options: context lines (-A/-B/-C), file glob filtering (e.g. "*.ts", "src/**/*.ts"), and output modes (content, files_with_matches, count).

  Connector

• restore_version

  WebSlop

  DESTRUCTIVE: Restore an app to a previous version using git reset --hard. This permanently overwrites all current files with the state from the specified commit — any changes made after that commit will be lost and CANNOT be recovered. You MUST confirm with the user before calling this tool. Use list_versions to show the user available versions first.

  Connector

• verify_code_integrity

  Solentic

  Verify the code running on Blueprint servers. Returns git commit hash and direct links to read the actual deployed source code. Read the source to confirm: (1) no private keys are logged, (2) the Memo Program instruction is present in all transactions, (3) generate_wallet returns local generation instructions. Don't trust — read the code yourself via the source endpoints.

  Connector

• mental-health-get-test

  Well-Being Check

  Loads one supported self-assessment into the widget by slug. Use `gad7` for anxiety screening, `phq9` for depression screening, and `who5` for general well-being screening when the user wants to take one of those assessments.

  Connector

• get-reward-policy

  Crinkl Commerce MCP

  Get the current Crinkl reward policy. Returns the active policy version, SHA-256 policy hash, reward parameters (referral bonuses, base points), category multipliers (grocery 1.5×, dining 1.5×, etc.), $CRINKL token holding tier multipliers, and the latest reserve checkpoint (BTC price, sats/receipt, coverage ratio). All values are deterministic — the policy hash covers the full parameter set. Updated daily.

  Connector

• deploy_app

  mcp

  Deploys an app to a VM and exposes it at a public https://<name>-<id>.redu.cloud URL. The container is built ON the VM. PREREQS — run check_deploy_prerequisites first for network_id + keypair_name, then plan_deploy for cost approval. Source can be git repo or prepare_upload source_token. PORT must be the real app listen port. To wire a DB, pass database:'managed' (dedicated managed datastore VM on the same private network, reused on same-name redeploy) or database:'single_vm' for Postgres on the app VM. Choose db_engine ('postgres' default; 'mysql'/'mariadb' for WordPress/Matomo/LAMP, managed only). For WordPress/WooCommerce cluster intent, do not use generic stateless deploy: pass app_profile, cluster_target:true, database:'managed', db_engine:'mariadb' or 'mysql', cluster_media_mode:'media_space', and either media_space_id or create_media_space:true. Redu mounts the media space into wp-content/uploads and refuses unsafe local uploads. Build+provision takes minutes; poll list_deployments/get_deployment.

  Connector

• deploy

  mcp

  Deploy an application to sota.io. The platform auto-detects your framework and builds a Docker image automatically: - Next.js: Detected via next.config.js/ts. Add output: 'standalone' to next.config for optimal builds. - Node.js: Detected via package.json with a "start" script. Works with Express, Fastify, Koa, Hapi, etc. - Python: Detected via requirements.txt or pyproject.toml. Works with Flask, FastAPI, Django. - Custom Dockerfile: If a Dockerfile exists in the project root, it takes priority over auto-detection. Use this for Go, Rust, Java, or any other language. The EXPOSE directive in the Dockerfile is used to detect the app port automatically. THREE WAYS to supply the source code — pick EXACTLY ONE: 1. **files** (inline source from AI): Pass a map of relative paths to UTF-8 text content. Best when you've just generated a small app in this conversation and want to deploy it without any filesystem step. Up to 200 files, 10 MB total. Include the framework manifest (package.json, requirements.txt, or Dockerfile) so auto-detection works. 2. **git_url** (clone a public repo): Pass an https://, git://, ssh://, or git@host:path URL. We shallow-clone it (--depth=1 --single-branch) on the server and deploy. Optional git_branch picks a non-default branch. Only public repos are supported in v1. Max 200 MB after clone. 3. **directory** (local filesystem): Pass an absolute path. Only works when the MCP client has filesystem access (Claude Code / CLI; not Claude.ai web). Defaults to the current working directory when omitted. IMPORTANT: Your app MUST listen on the PORT environment variable. For auto-detected frameworks (Next.js, Node.js, Python) PORT is 8080. For custom Dockerfiles, the port is auto-detected from the EXPOSE directive (e.g. EXPOSE 3000 sets PORT=3000). If no EXPOSE is found, it defaults to 8080. Every project includes a managed PostgreSQL 17 database. Six environment variables are auto-injected into your container — no manual database configuration needed: DATABASE_URL (full connection string), PGHOST, PGPORT, PGUSER, PGPASSWORD, and PGDATABASE. Libraries that follow libpq conventions (node-postgres, pgx, psycopg2, Django) pick up the PG* variables automatically with no configuration. If your app needs database migrations, run them on startup. Deployments use blue-green strategy for zero downtime. The old container keeps running until the new one passes health checks (60s timeout). Use get-logs to monitor build progress. Files matching .gitignore, .git/, node_modules/, .env, and .DS_Store are excluded from the archive.

  Connector

• project-create

  Planner (proof-of-done)

  Create a new project container for goals. Requires title and a unique slug (lowercase a-z, 0-9, hyphens). Optional: description, status (active/archived/paused, default active), tags, icon (emoji), repositoryPath, repositoryUrl, sourceRef ({type:"git",url,ref} — remote-research source), isDefault. Returns the created project with id, slug, and webUrl. Slug must be unique — duplicates are rejected with existing slugs list.

  Connector

• claim_issue

  retasc

  Claim one specific issue by id (instead of letting next_issue pick). Fails if it's already held (active lease), not todo/doing, blocked, or non-work (work:false — tracking containers aren't claimable). Returns a claim_token. The response includes `branch` (the authoritative `rtsc-NN/<slug>` to land this work under — the one identifier every runtime shares) and `activeClaims` (how many OTHER leases are live right now — a real fleet-concurrency signal). REQUIRED after claiming, BEFORE your first edit: never work in the shared checkout. Create an isolated git worktree for `branch` and work ONLY there — e.g. `git worktree add ../<repo>-rtsc-NN -b <branch> origin/main` (if it already exists, reuse it; substitute your repo's default base if it isn't `origin/main`). Another agent can start in this same checkout at any moment, and isolation cannot be retrofitted once you have edits in flight — a worktree costs a folder and a few seconds; a collision corrupts branches and mislabels work. Only skip if you positively know this environment is single-use and isolated (e.g. an ephemeral per-task container) — and even then a worktree is harmless. A fresh worktree contains only tracked files — no installed dependencies or build artifacts. Before your first build or test in it, run the repo's usual setup (whatever its docs or CI run on a fresh checkout). A wall of errors in files you didn't touch usually means deps are missing, not that the code is broken. Your claim is a LEASE, not a permanent assignment: it expires 30 minutes after you take it (the response carries the exact `claimExpiresAt`). ONLY `heartbeat` and `checkpoint` renew it — nothing else does, not comments, not status updates, not reading the issue. Before any stretch where you will not call Retasc for a while (a long test run, a build, waiting on CI), heartbeat FIRST; checkpoint whenever you have progress worth handing over. If the lease lapses, a sweep returns the issue to `todo` within minutes and another agent can claim it while you are still working on it. If this call is refused for billing (BILLING_INACTIVE or MONTHLY_CAP_REACHED), the response carries `tellHuman`, a ready-to-relay sentence: PASS IT TO YOUR HUMAN. You cannot fix it yourself — only an owner can — so do NOT absorb the refusal and continue with read-only work as if the session were healthy; every value-bearing write will keep failing until they act.

  Connector

• prepare_upload

  mcp

  Returns the LOCAL shell commands to package your working directory and upload it for an upload-mode deploy (no git, no PAT). Run them in the user's terminal, capture `source_token` from the upload's JSON response, then call deploy_app with that source_token (omit repo). The upload authenticates AUTOMATICALLY with a short-lived ticket minted from your MCP credential — NO API key needed in the command and nothing secret is printed (it falls back to needing $REDU_API_KEY only if minting is unavailable). Excludes node_modules/.git/.venv/build output and .env by default; honors .gitignore when is_git_repo=true.

  Connector

• build_and_deploy

  vibedeploy

  Run a build inside a hardened one-shot pod against the site's editable source tree (write source first via write_source_files / list_source_files autoPromote), then atomically swap the build output into the live dist. Reuses the same build pod the GitProject git-deploy flow uses, so the same isolation guarantees apply: no SA token, no DB/Vault reach, NetworkPolicy-restricted egress. The first run writes the chosen buildCommand/outputDir into Site.sourceManifest; subsequent calls can omit those fields.

  Connector

• create-project

  mcp

  Create a new project on sota.io. Each project automatically provisions: (1) a managed PostgreSQL 17 database accessible via the DATABASE_URL environment variable (auto-injected, no configuration needed), (2) PgBouncer connection pooling (pool size 20, max 100 clients), (3) automatic daily database backups with 7-day retention, (4) a live URL at https://{slug}.sota.io with automatic HTTPS via Let's Encrypt. The project slug is auto-generated from the name (lowercase, hyphens, max 63 chars) and is immutable after creation. Supported frameworks: Next.js, Node.js (Express/Fastify/Koa), Python (Flask/FastAPI/Django), or any language via custom Dockerfile. You can also add up to 5 custom domains per project with automatic HTTPS (via API: POST /v1/projects/:id/domains with {domain: "yourdomain.com"}). DNS: A record to 23.88.45.28 for apex domains, CNAME to {slug}.sota.io for subdomains. Optionally associate the project with a public git repository at create-time by passing `git_url` (and optional `git_branch`). The association is informational — it shows up in the dashboard and the `sota deploy --git` CLI flag can default to it — but does NOT enable auto-deploy-on-push yet.

  Connector

• project-update

  Planner — provable "done" for AI agents

  Partial update of a project — only the fields you pass are changed; omitted fields are untouched. Updatable: title, slug, description, status (active/archived/paused), tags, icon (emoji), repositoryPath, repositoryUrl, sourceRef ({type:"git",url,ref} — remote-research source), clearSourceRef (true → сбрасывает Project.sourceRef в null; несовместим с sourceRef), isDefault, redTeamMode (optional/advisory/required), evidenceJudgeMode (off/optional/required), allowedGoalMode (any/grove_only), fmeaGenerationMode (off/on), escalationResolveTimeout (off/1h/4h/24h), escalationAutoResolveBy (none/planner-agent), escalationAutoResolveStrategy (suggest-edit/reject-default). Returns the updated project with all fields.

  Connector