Codevira MCP
Codevira
One memory layer for every AI coding tool you use. Switch between Claude Code, Cursor, Windsurf, and Antigravity without losing context, decisions, or progress.
Built for solo developers working on local projects with AI agents. Codevira gives every AI tool you use access to the same persistent project memory — so you stop re-explaining your codebase every session, stop losing carefully-made decisions, and stop burning tokens on re-discovery.
Works with: Claude Code · Claude Desktop · Cursor · Windsurf · Google Antigravity · any MCP-compatible AI tool
The Problem (Four Pains Codevira Solves)
If you code with AI agents on a project longer than a week, you've felt all of these:
1. Re-explaining your codebase every session
Every new chat starts from zero. The AI doesn't know your architecture, your conventions, your "we don't do it that way" decisions. You waste the first 10 minutes (and thousands of tokens) catching it up — only to do it again tomorrow.
2. AI undoing your careful decisions
Last week you debugged a tricky retry policy for 3 hours. Today's AI session refactors it to a simpler version because it has no idea why the complexity exists. Now it's broken again.
3. Cross-tool amnesia
You started planning in Claude Code. Switched to Cursor for autocomplete. Opened Antigravity to run tests. Three different agents, three different blind copies of your project state. Nothing carries over.
4. Token budget burned on re-discovery
Your AI agent reads the same 12 files every session before doing any actual work. You're paying API costs for the same lookups, over and over.
Codevira is a persistent memory layer that fixes all four — for every AI tool, on every project, on your local machine.
Related MCP server: Tages
How It Works
Codevira is a Model Context Protocol server that runs locally and gives any AI tool a structured, queryable memory of your codebase:
Capability | What it means for you |
Zero-config setup |
|
Cross-tool continuity | One |
Decision protection |
|
Context graph | Every source file has a node: role, rules, dependencies, stability, blast radius. AI calls |
Function-level call graph |
|
Semantic code search | Natural-language search across your codebase ( |
Roadmap + changesets | Multi-file work tracked atomically; sessions resume cleanly after interruption |
Adaptive learning | Tracks which past decisions panned out — gives confidence scores and surfaces patterns |
Cross-project memory | Learned preferences sync across all your local projects via |
Auto-init on first call | No |
Token-efficient by design
Codevira is built around the principle that AI agent context windows are precious. Tools return summaries by default with opt-in full data:
get_node(path)— ~100 tokens by default (counts + flags). Passfull=truefor the entire rules array.get_impact(path)— 10 affected files. Passsummary_only=truefor just counts (~80 tokens) before deciding to dig deeper.search_codebase(query)— file/symbol pointers only. Passinclude_content=trueto inline source.search_decisions(query)— 5 truncated matches. Passfull=truefor verbatim text.
The agent always asks for what it needs, in the size it needs.
Quick Start
1. Install
# Recommended: global install via pipx (isolated, works everywhere)
pipx install codevira
# Alternative: pip install
pip install codeviraInstalls the full toolkit (23 AI-facing MCP tools + 12 admin/CLI tools) out of the box. Semantic search downloads a ~90MB embedding model on first use.
2. Register with your AI tools
codevira registerThis one-time global command injects Codevira's MCP config into all detected AI tools — Claude Code, Cursor, Windsurf, Claude Desktop, and Google Antigravity. Run it from anywhere; no project directory needed.
3. Start using
Open any project in your AI tool. On the first MCP tool call, Codevira auto-initializes:
Detects language, source directories, and file extensions from project markers
Creates the context graph and roadmap
Installs a
post-commitgit hook for automatic reindexing
No explicit codevira init needed — everything happens on demand.
Note:
codevira initis still available for explicit per-project setup with custom settings.
4. Verify
Ask your AI agent to call get_roadmap() — it should return your current phase and next action.
Note: Restart your AI tool after running
codevira registerto pick up the new MCP config.
Customizing what's indexed
Codevira tries to auto-detect your project's source layout, but monorepo or non-standard layouts sometimes slip through — you'll notice when codevira index --full reports 0 chunks indexed and prints a hint pointing you here.
cd your-project
codevira configureScans your project (gitignore-aware), shows discovered directories and extensions with file counts, and lets you pick which to watch via a numbered-list prompt. It writes your choices back to .codevira/config.yaml and offers to rebuild the index.
Non-interactive (useful in scripts or CI):
codevira configure --dirs src,packages,apps --extensions .py,.ts,.tsx --no-reindexAfter changing watched directories, restart your AI tool — running watchers snapshot the dir set at boot.
Manual config (only if auto-inject didn't detect your tool)
Codevira supports two transports. Use the right one for your client:
Client | Transport | Config file |
Claude Desktop (app) | stdio |
|
Claude Code (CLI) | stdio or HTTP |
|
Cursor | stdio |
|
Windsurf | stdio |
|
Google Antigravity | stdio |
|
Stdio transport — Claude Desktop, Cursor, Windsurf (.claude/settings.json / .cursor/mcp.json / .windsurf/mcp.json):
{
"mcpServers": {
"codevira": {
"command": "codevira",
"args": [],
"cwd": "/path/to/your-project"
}
}
}Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"codevira": {
"command": "/path/to/codevira",
"args": ["--project-dir", "/path/to/your-project"]
}
}
}Tip: find the full binary path with
which codevira
HTTP/HTTPS transport — Preview in v1.7, single-project only. The HTTP server binds to one project at startup and cannot switch contexts per request. Multi-project HTTPS is planned for v1.8. For multi-project work today, use stdio via codevira register (above).
First start the HTTP server in a terminal:
codevira serve --port 7007 --project-dir /path/to/your-project
# For HTTPS (required by some clients):
codevira serve --https --port 7443 --project-dir /path/to/your-projectThen register the URL:
{
"mcpServers": {
"codevira": {
"url": "https://localhost:7443/mcp"
}
}
}HTTPS note: Claude Code uses Node.js, which requires a trusted CA for HTTPS. Run once to trust the mkcert CA:
brew install mkcert && mkcert -install launchctl setenv NODE_EXTRA_CA_CERTS "$(mkcert -CAROOT)/rootCA.pem" echo 'export NODE_EXTRA_CA_CERTS="$(mkcert -CAROOT)/rootCA.pem"' >> ~/.zshrcThen restart Claude Code.
Auto-start on login (macOS):
codevira serve --install-service # start server automatically on login
codevira serve --uninstall-service # remove auto-startGoogle Antigravity (~/.gemini/antigravity/mcp_config.json):
{
"mcpServers": {
"codevira": {
"$typeName": "exa.cascade_plugins_pb.CascadePluginCommandTemplate",
"command": "codevira",
"args": []
}
}
}Codevira data layout (v1.6)
~/.codevira/ <- global Codevira home
├── global.db <- cross-project intelligence
├── projects/
│ └── <project-key>/ <- per-project data (keyed by path)
│ ├── config.yaml
│ ├── metadata.json
│ ├── graph/
│ │ ├── graph.db
│ │ └── changesets/
│ ├── codeindex/ <- semantic search (optional)
│ └── logs/
└── certs/ <- HTTPS certs (if using --https)Legacy
.codevira/directories inside project repos are auto-migrated to centralized storage on first server start.
Configuration
Each project has a config.yaml at ~/.codevira/projects/<project-key>/config.yaml. It's auto-generated on first use with sensible defaults, but you can edit it to customize what Codevira indexes:
project:
name: my-project
language: python
collection_name: my_project
# Which directories to scan for source files
watched_dirs:
- src
- tests
- scripts
# Which file extensions count as "source" for indexing + change detection
file_extensions:
- .py
- .ts
- .tsx
# Directories to skip even if inside watched_dirs
skip_dirs:
- node_modules
- .venv
- __pycache__
- dist
- build
logs:
# 0 = keep sessions/decisions forever (default).
# Only set > 0 if you have privacy reasons to time-bound history.
retention_days: 0Common gotchas:
file_extensionsmust be a proper YAML list — each extension on its own line. This is wrong:file_extensions: - .py, .md, .html # ❌ one item containing commas, not three extensionsThis is correct:
file_extensions: - .py - .md - .htmlOr inline:
file_extensions: [.py, .md, .html]file_extensionsis intended for source code (Python, TypeScript, Go, Rust, etc.). Codevira uses tree-sitter AST parsing — putting.mdor.htmlhere may produce malformed graph nodes since tree-sitter parsers for those languages are different.Files are only scanned if they live inside
watched_dirs. Adding an extension alone isn't enough — make sure the directory is listed too.
After editing the config, run codevira index --full to rebuild the graph from scratch, or codevira index for incremental changes.
Uninstall / Reset
codevira clean # remove global data + IDE configs + launchd service
codevira clean --all # also remove per-project artifacts
codevira clean --dry-run # preview what would be removedHow It Works
Setup Flow
flowchart LR
A["pipx install codevira"] --> B["codevira register"]
B --> C["Open project in\nClaude Code / Cursor /\nWindsurf / Antigravity"]
C --> D["First MCP tool call\ntriggers auto-init"]
D --> E["✓ Config written\n✓ Graph built\n✓ Roadmap created\n✓ Ready"]Agent Session Lifecycle
flowchart TB
Start([Start Session])
subgraph "Orientation (single call)"
A["get_session_context()\nroadmap + changesets +\ndecisions + global intelligence"]
end
subgraph "Work"
B[get_node / get_impact\nbefore touching files]
C[Plan + Implement + Test]
D[refresh_index\nafter changes]
end
subgraph "Session End"
E[update_node — record changes]
F[write_session_log — decisions]
G[update_next_action — handoff]
end
Start --> A
A --> B
B --> C
C --> D
D --> E
E --> F
F --> GArchitecture
flowchart TB
A[Source Code\n15+ languages]
subgraph "Indexing Pipeline"
B[Tree-sitter AST Parser]
C[Function / Class / Call Extraction]
D[Background File Watcher\nauto-reindex on save]
end
subgraph "Centralized Storage — ~/.codevira/"
E[(Context Graph + Call Graph\nSQLite DB)]
F[(Semantic Index\nChromaDB — optional)]
G[(Global Memory\ncross-project intelligence)]
H[(Session Logs + Decisions\nsearchable history)]
end
subgraph "Adaptive Learning"
I[Outcome Tracking]
J[Rule Inference]
K[Preference Learning]
end
subgraph "MCP Server"
L[36 Tools + 5 Prompts\nstdio or HTTP transport]
end
M[AI Coding Agent\nClaude Code · Cursor · Windsurf · Antigravity]
A --> B
B --> C
C --> E
C --> F
D --> B
E --> L
F --> L
G --> L
H --> L
I --> G
J --> G
K --> G
E --> I
L --> MSession Protocol
Every agent session follows a simple protocol. Set it up once in your agent's system prompt — then your agents handle the rest.
Session start (mandatory):
list_open_changesets() -> resume any unfinished work first
get_roadmap() -> current phase, next action
search_decisions("topic") -> check what's already been decided
get_node("src/service.py") -> read rules before touching a file
get_impact("src/service.py") -> check blast radiusSession end (mandatory):
complete_changeset(id, decisions=[...])
update_node(file_path, changes)
update_next_action("what the next agent should do")
write_session_log(...)This loop keeps every session fast, focused, and resumable.
MCP Tools + 5 Prompts
23 tools exposed to AI agents (token-optimized, summary-first). The remaining 12 tools are admin/dashboard tools that work via dispatch but aren't advertised in list_tools() — humans access them via the CLI or via specific MCP prompts. Tools marked (admin) below.
Graph Tools
Tool | Description |
| Summary by default (counts + flags); |
| Blast radius — |
| Append rules, connections, key_functions |
| Function-level: callers, callees, tests, dependents, symbols |
| Bulk node listing — agents should use targeted queries instead |
| Register a new file (auto-generated by refresh_graph) |
| Auto-generate stubs (background/automatic) |
| Background reindex (fire-and-forget) |
| Mermaid/DOT export — large dump |
| PR review — use |
| Risk scoring — use |
| Complexity dashboard |
Roadmap Tools
Tool | Description |
| Current phase, next action, open changesets |
| Full details of any phase by number |
| Set what the next agent should do |
| Mark phase in_progress / blocked |
| Queue new upcoming work |
| Mark done, auto-advance to next |
| Move a phase to the deferred list |
| Full history with all decisions inline |
Changeset Tools
Tool | Description |
| All in-progress changesets |
| Open a multi-file changeset |
| Close and record decisions |
| Mid-session checkpoint |
Search Tools
Tool | Description |
| Semantic search — pointers only by default |
| Past decisions (default 5, truncated context) |
| Recent decisions touching a file (default 5) |
| Write structured session record |
Adaptive Learning Tools
Tool | Description |
| THE main "catch me up" call — start every session here (~800 tokens) |
| Outcome-based reliability scores |
| Already in get_session_context |
| Already in get_session_context |
| Dashboard metric — use |
Code Reader Tools
Tool | Description |
| All public symbols, signatures, line numbers (Python, TypeScript, Go, Rust) |
| Full source of one function or class |
Playbook Tool
Tool | Description |
| Curated rules for: |
MCP Workflow Prompts (v1.5)
Prompt | Description |
| Staged diff + blast radius + risk score |
| Symptom -> affected files -> call chain -> hypothesis |
| Full project context catch-up for new sessions |
| Test coverage gaps + high-risk functions before commit |
| Module map + hotspots + dependency summary |
Language Support
Feature | Python | TypeScript | Go | Rust | 12+ Others |
Context graph + blast radius | Y | Y | Y | Y | Y |
Semantic code search | Y | Y | Y | Y | Y |
Function-level call graph | Y | Y | Y | Y | |
| Y | Y | Y | Y | |
AST-based chunking | Y | Y | Y | Y | |
Auto-generated graph stubs | Y | Y | Y | Y | |
Roadmap + changesets | Y | Y | Y | Y | Y |
Session logs + decision search | Y | Y | Y | Y | Y |
Supported languages: Python, TypeScript, JavaScript, Go, Rust, Java, Kotlin, C#, Ruby, PHP, C, C++, Swift, Solidity, Vue.
Requirements
Python 3.10+
~500MB install (includes ChromaDB + sentence-transformers for semantic search)
~90MB model download on first
search_codebase()call
pip install codevira includes the full toolkit out of the box — graph, roadmap, changesets, code reader, learning, call graph, and semantic search.
Minimal install (no semantic search)
If you want to skip the ML stack and use only graph-based tools (semantic search disabled), install without the search deps:
pip install codevira --no-deps
pip install pyyaml mcp watchdog tree-sitter tree-sitter-language-pack rich uvicorn starlette pathspecThe search_codebase tool will be hidden from your AI agent; all other tools work normally.
Background
Want to understand the full story behind why this was built, the design decisions, what didn't work, and how it compares to other tools in the ecosystem?
Read the full write-up: How We Cut AI Coding Agent Token Usage by 92%
Contributing
Contributions are welcome. Read CONTRIBUTING.md for the full guide.
Reporting a bug? Open a bug report Requesting a feature? Open a feature request Found a security issue? Read SECURITY.md — please don't use public issues for vulnerabilities.
Testing a release candidate locally? See docs/local-pypi-https.md for setting up a Docker-based HTTPS PyPI registry that mirrors the real PyPI install flow without touching public PyPI.
FAQ
Common questions about setup, usage, architecture, and troubleshooting — see FAQ.md.
Roadmap
See what's built, what's next, and the long-term vision — see ROADMAP.md.
Star History
If Codevira saves you tokens or sanity, a star helps other developers find it. Tracking growth keeps me focused on what's working.
License
MIT — free to use, modify, and distribute.
Available Tools
36 toolsadd_phaseA
Add a new upcoming phase to the roadmap. Call when you identify new work during a session — gaps, refactors, follow-ups. High-priority phases are inserted at the front of the queue.
Inputs: name (required), phase (required), description (required), files, +3 more.
apply_skill_outcomeA
v3.1.0 M3: Manually record one outcome for a skill — success or failure. Reinforces the reinforcement loop (resets consecutive_failures on success; auto-archives at 5 consecutive failures unless do_not_revert=True). The canonical signal in M5+ comes from outcomes_writer.py (git-derived, not agent-self-reported); this tool is the manual override.
Inputs: success (required), skill_id (required).
bulk_import_phasesA
v2.1.2 Item 29: backfill multiple historical phases at once. Each item: {number, name, status?='done', completed_at?, key_decisions?, git_ref?, description?}. Idempotent. Useful for adopting codevira on a project that already shipped N phases in git.
Inputs: phases (required).
check_conflictARead-onlyIdempotent
Check whether a proposed decision contradicts any do_not_revert=True decision OR duplicates an existing one. A NEGATED restatement is always a conflict, never a duplicate: 'never do X' and 'do X' differ by one token and score as near-identical text, so they are separated explicitly. Returns {status: novel|duplicate|conflict, conflicts, duplicates}. Call BEFORE record_decision to surface conflicts proactively (record_decision also runs this internally and surfaces _conflict_warning unless force=true).
Inputs: decision_text (required), file_path.
complete_phaseA
Mark the current phase as complete and advance to the next upcoming phase. Records key_decisions permanently. Requires phase_number to match current phase (safety check). v2.1.2 Item 10: pass backfill=True + completed_at='YYYY-MM-DD' to retroactively mark a historical phase done without advancing the queue. v2.1.2 Item 12: pass git_ref to link a commit sha or PR ref to the completion.
Inputs: phase_number (required), key_decisions (required), git_ref, backfill, +1 more.
defer_phaseA
Move an upcoming phase to the deferred list. Use when priorities shift or a phase depends on unavailable work.
Inputs: reason (required), phase_number (required).
expandARead-onlyIdempotent
E1 (Phase 19): fetch FULL decision records by ID — the expand path for the summary-first search_decisions / list_decisions defaults. Scan the cheap compact rows, then pass the IDs you care about here for complete text + context + origin. Returns {requested, count, decisions, not_found}; never raises on unknown IDs.
Inputs: ids (required).
get_historyARead-onlyIdempotent
Get recent decisions touching a file. Default: 5 with truncated context (~500 tokens). Pass full=true for untruncated text. Ordered by most recent first.
Inputs: file_path (required), full, limit.
get_impactARead-onlyIdempotent
Get the blast radius for a file before modifying it. Default: returns up to 10 affected files + counts (~400 tokens). Pass summary_only=true for just counts (blast_radius, protected_count, high_stability_count) — ~80 tokens, perfect for gate checks. ALWAYS call before modifying any file.
Inputs: file_path (required), limit, summary_only.
get_nodeARead-onlyIdempotent
Get the context graph node for a file. Returns a SUMMARY by default (role, layer, stability, rules_count, deps_count, stale flag) — ~100 tokens. Pass full=true for the complete rules/dependencies/key_functions arrays. Call this INSTEAD of reading the source file.
Inputs: file_path (required), full.
get_phaseARead-onlyIdempotent
Get full details of any phase by number — completed, current, or upcoming.
Inputs: phase_number (required).
get_playbookARead-onlyIdempotent
Get curated architectural rules for a specific task type. Returns only the 2-3 relevant rule files — not all of them. Valid task types: add_tool | add_service | add_schema | debug_pipeline | commit | write_test
Inputs: task_type (required).
get_roadmapARead-onlyIdempotent
Get current project state: phase number, name, status, next action, and upcoming phases. Call at the START of every session.
Inputs: no named parameters.
get_session_contextARead-onlyIdempotent
Single 'catch me up' call for cross-tool continuity. Returns current roadmap phase, recent decisions with confidence, learned preferences, and active rules — everything a new session needs. Call this at the START of every session instead of multiple separate calls. recent_decisions is ranked by recency x outcome-confidence (a decision git watched survive outranks one nothing has tested); reverted and outdated ones are hidden entirely. A decision whose file has CHANGED since it was recorded carries 'needs_review': true plus a 'review_hint' — reaffirm_decision if it still holds, supersede_decision or mark_decision_outdated if it does not. Works seamlessly across AI tools: Cursor, Claude Code, Antigravity.
Inputs: no named parameters.
get_skillARead-onlyIdempotent
v3.1.0 M3: Composite-ranked search over active skills. score = 0.5 × BM25_norm + 0.3 × tag_jaccard + 0.2 × recency_decay (τ=30d, never-used skills score 0 recency). Returns hits with score_breakdown for debuggability. Pass file_path to filter skills whose trigger file_patterns don't match.
Inputs: query (required), top_k, file_path.
get_working_contextARead-onlyIdempotent
v3.1.0 M2: Compact markdown rendering of the top working-memory entries for ReAct-loop injection. Returns {markdown, entries, count}. Capped at ~150 tokens of output (entries truncated at 120 chars each).
Inputs: top_k.
list_decisionsARead-onlyIdempotent
v2.1.2 Item 11: enumerate decisions with filters (since_date, file_pattern, protected_only, session_id, tags). Closes the gap that 'codevira can remember things across sessions, but can't list what it remembers.' Default (E1): compact rows — one-line decision summary + key fields; full=true (or CODEVIRA_DECISION_DETAIL=full) for untruncated records, expand(ids=[...]) to fetch specific decisions in full.
Inputs: full, tags, limit, session_id, +5 more.
list_skillsARead-onlyIdempotent
v3.1.0 M3: Filtered list of skills. status='active' (default) returns the daily-driver set; 'all' returns every state; any other value filters to that one state. tags filter is set intersection.
Inputs: tags, limit, source, status.
list_tagsARead-onlyIdempotent
v2.1.2 Item 27: enumerate all tags in the project with decision counts. Useful for discovery — 'what categories of decisions do we track?'
Inputs: no named parameters.
mark_decision_outdatedA
v3.7.0 staleness read-side: tombstone a decision as OUTDATED so it stops surfacing in get_session_context / search_decisions / list_decisions — without deleting it. Use when a decision is simply no longer true and has NO successor (for a replacement, use supersede_decision to preserve lineage). Reversible via set_decision_flag(is_outdated=false). Writes one amendment to .codevira/decisions.jsonl; audit preserved.
Inputs: decision_id (required), force, reason.
origin_ofBRead-onlyIdempotent
v3.1.0 M7: Return the M1 origin block attached to a decision ({ide, agent_model, host_hash, ts}) + protection / supersession metadata. Always available regardless of the handshake flag.
Inputs: decision_id (required).
promote_skill_to_playbookA
v3.1.0 M3: Write the skill's procedure as a playbook markdown file at .codevira/playbooks/<task_type>/.md. Refuses on existing file unless force=True so hand-written playbooks aren't clobbered. After promotion the procedure is also discoverable via get_playbook(task_type).
Inputs: skill_id (required), task_type (required), name, force.
query_graphARead-onlyIdempotent
Query the function-level call graph. Find callers, callees, tests, or dependents for a specific symbol. Use query_type='symbols' to list all functions in a file.
Inputs: file_path (required), symbol, query_type.
reaffirm_decisionA
v3.2.0: refresh a do_not_revert decision's soft-expire clock. Long-lived locked decisions can grow stale, so search and list output carry 'dnr_soft_expired' and 'dnr_age_days' on every do_not_revert decision — 180 days by default, override via CODEVIRA_DNR_SOFT_EXPIRE_DAYS (0 disables). The lock never auto-flips; the flag is advice. Call this on a soft-expired decision that is still load-bearing — it appends a single 'reaffirmed_at' amendment to .codevira/decisions.jsonl and resets the clock. For semantic rewrites use supersede_decision; for flipping the flag use set_decision_flag.
Inputs: decision_id (required).
record_decisionA
Record one architectural decision. Set do_not_revert=true to lock it across sessions and IDEs. Returns {decision_id, session_id}. To change it later use supersede_decision (preserves the audit trail) or set_decision_flag (toggle do_not_revert / tags).
Inputs: decision (required), tags, force, symbol, +6 more.
record_skillA
v3.1.0 M3: Author a new skill in the canonical store (.codevira/skills.jsonl). Skills encode 'how to do X in this project' as markdown procedures. Calls check_conflict against the SKILLS corpus before writing; near-duplicate warnings can be overridden via force=True. Use supersede_skill to version an existing skill, or promote_skill_to_playbook to promote a skill into the existing playbook system.
Inputs: name (required), procedure (required), force, source, +3 more.
search_decisionsARead-onlyIdempotent
Search past decisions across sessions and roadmap phases. Keyword search over an FTS5/BM25 index (Porter-stemmed) — NO semantic/vector matching, so recall depends on sharing keywords with the stored decision; for a concept with no shared words, browse list_decisions or list_tags instead. Default (E1): summary-first rows — {id, decision (one-line ≤140), file_path, do_not_revert, tags, score}, dropping per-row snippet/origin. Pass full=true for untruncated rows, expand(ids=[...]) to fetch specific decisions in full, or summary_only=true for a ~70%-smaller {id, summary, score} payload. Answers 'has anyone decided this before?'
Inputs: query (required), full, limit, session_id, +2 more.
set_decision_flagA
v3.0.0 lightweight flag/tag update for an existing decision. Use this when you only need to toggle do_not_revert or correct a tag list — avoids supersede_decision's mandatory rewrite of the decision text + reason. Writes a single amendment record to .codevira/decisions.jsonl. For semantic rewrites use supersede_decision instead.
Inputs: decision_id (required), tags, force, is_outdated, +1 more.
supersede_decisionA
v2.1.2 Item 26: retire old_id and link to a replacement. Writes the new decision with [supersedes #<old_id>: <reason>] prefix, sets the old row as superseded. Default-hidden in search / list (pass include_superseded=true to opt back in).
Inputs: old_id (required), reason (required), new_decision (required), tags, +3 more.
supersede_skillA
v3.1.0 M3: Version a skill. Writes a new skill that supersedes old_id; amendment-marks the old as 'superseded' with a backref. Triggers inherit from the old skill when not supplied. The old skill no longer surfaces in search after this; it's still retrievable via list_skills(status='superseded') for audit.
Inputs: name (required), old_id (required), procedure (required), reason, +3 more.
update_next_actionA
Update the roadmap's next_action field. Call at session end.
Inputs: next_action (required).
update_phase_statusA
Update the current phase status: pending | in_progress | blocked. Call when starting work on a phase (in_progress) or when blocked.
Inputs: status (required), blocker, started.
working_addA
v3.1.0 M2: Append one observation or goal to working memory (intra-session, bounded, decay-scored scratchpad in .codevira-cache/working.jsonl). 'observation' = a fact the agent saw (file edited, error message, command output). 'goal' = what the agent is currently trying to accomplish. Use working_promote to move an entry to long-term memory (decision/skill/playbook) when it earns its keep.
Inputs: content (required), kind, links, confidence, +2 more.
working_getARead-onlyIdempotent
v3.1.0 M2: Top-K live working-memory entries by decay score (importance × exp(-Δt_hours / 6) + 0.5 × access_count). Filters by kind / session_id. Tombstoned (evicted or promoted) entries are excluded.
Inputs: kind, top_k, session_id.
working_promoteB
v3.1.0 M2: Promote a working-memory entry to long-term memory and tombstone the source. to='decision' is the fully wired path (calls check_conflict first; force=true overrides). to='skill' and to='playbook' are reserved for M3+; the call returns {deferred: true, milestone: ...} until those stores ship.
Inputs: entry_id (required), to, tags, force, +3 more.
write_session_logA
Write a structured session log to .agents/logs/YYYY-MM-DD/. Called by the Documenter at the end of every session. Feeds search_decisions() with institutional memory.
Inputs: task (required), phase (required), decisions (required), next_steps (required), +4 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 36 tools
Each tool targets a distinct operation and resource: roadmap phases, decision records, working memory, skills, and graph queries are cleanly separated. Even related tools like search_decisions vs list_decisions and get_node vs query_graph are clearly differentiated by their descriptions and intended use cases.
The overwhelming majority of tools follow snake_case verb_noun naming like add_phase, get_roadmap, search_decisions, and record_skill. Minor deviations such as working_add, working_get, expand, and origin_of break the pattern slightly but are still readable and not chaotic.
36 tools is well beyond the 25+ threshold that should be considered too many for a coherent MCP surface. While the tools cover multiple subdomains, the sheer number makes the set heavy for an agent to navigate and weigh in context.
The tool surface provides deep lifecycle coverage for its apparent domain: roadmap phases, decision records, skills, working memory, and file context graph. There are few obvious missing operations — decisions can be recorded, searched, superseded, flagged, and outdated; skills can be created, listed, promoted, and managed; and phases have create, read, update, complete, and bulk-import operations.
Maintenance
Related MCP Connectors
Long-term memory for AI coding agents: durable project facts, recalled by every MCP client.
Persistent engineering memory designed to follow developers across coding agents.
Hosted MCP memory for coding agents: persistent across sessions, editable markdown, team sharing.
Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP).
Related MCP Servers
- FlicenseNot gradedqualityNot gradedmaintenanceAn MCP server that provides persistent project context, workflow management, and knowledge capture for AI coding agents. It enables agents to maintain structured memory across sessions by tracking project profiles, conventions, skills, and technical debt.7-
- AlicenseNot gradedqualityBmaintenanceEnables AI coding agents to maintain persistent, cross-session memory of codebase architecture, naming conventions, and decisions through MCP tools. Eliminates repetitive project re-explanation by automatically injecting stored context into every session with local-first SQLite storage and optional team sharing capabilities.4MIT
- AlicenseNot gradedqualityCmaintenanceMCP memory server for AI coding agents to remember decisions, patterns, and bugs between sessions. Provides persistent memory with 37 MCP tools, multi-session coordination, and token-efficient recall.95 npm12MIT
- AlicenseBqualityBmaintenanceA shared, persistent MCP memory server for coding agents that enables storing and retrieving project decisions and context across different tools like Claude Code, Codex, and Cursor using semantic vector search.8MIT