RuleStack

Configs

Stacks

Compare

Diff

RuleStack

Configs

Stacks

Compare

Diff

Read API

RuleStack

Configs

Stacks

Compare

Diff

Read API

Configs/justrach/codedb/diff

Two files, one repository

justrach/codedb ships 3 formats across 3 indexed files. The question worth asking is whether the second one says anything the first does not.

CompareAGENTS.md ↔ CLAUDE.mdAGENTS.md ↔ Copilot instructionsCLAUDE.md ↔ Copilot instructions
A · AGENTS.md · 297 wordsB · CLAUDE.md · 244 words
What each file covers, counted
DimensionSharedOnly in AOnly in BOverlap
Sections0560%
Commands11320%
Section tags12317%

What each file covers

Sections

0 shared · 5 only in A · 6 only in B
  • − codedb Agent Guidelines
  • − What codedb is (and isn't)
  • − Review guidelines
  • − Pre-merge verification
  • − Security-sensitive areas
  • + codedb — Agent Instructions
  • + Project
  • + Rules
  • + Filing Issues
  • + Test Style
  • + Code Style

Commands

1 shared · 1 only in A · 3 only in B
  • − python3 scripts/e2e_mcp_test.py \
  • + zig build test-index
  • + zig build test 2>&1 | grep "issue-XX"
  • + gh issue create
  •   zig build test

Section tags

1 shared · 2 only in A · 3 only in B
  • − git-pr
  • − security
  • + test
  • + code-style
  • + do-not
  •   agent-behaviour

Line diff

+28 added−36 removed11 unchanged23.4% identical
justrach/codedb · AGENTS.md
@@ −1 @@
1# codedb Agent Guidelines
2 
3## What codedb is (and isn't)
4 
5codedb is a **code-intelligence and context tool, not an editor.** It exists to
6help agents *find and understand* code — structural search, symbol/caller
7lookup, dependency graph, outlines, and task-shaped context — so they can edit
8with their own native tools. codedb has **no edit capability**: the old
9`codedb_edit` fallback (and the HTTP `POST /edit` endpoint, the post-edit
10linter, and `codedb_diagnostics`) was removed because MCP edit calls bypass the
11client's own undo/rewind tracking. Keep that framing consistent across tool
12descriptions, the MCP `initialize` instructions, the README, and these docs.
13 
14## Review guidelines
15 
16- Flag any security issues: injection, file traversal, untrusted input, secret exposure
17- Verify that sensitive files (.env, .pem, .key, credentials) are excluded from indexing AND search
18- Check that telemetry behavior matches documentation claims
19- Flag any regression in benchmark-critical paths (threshold: 10%)
20- Treat P1 issues as merge-blocking
21- Verify new language parsers handle malformed input gracefully (braces in strings, unterminated comments)
22- Check that installer scripts don't execute untrusted code or skip verification
23 
24## Pre-merge verification
25 
26Run these before merging any MCP-related change:
27 
28```bash
29zig build test # unit tests
30python3 scripts/e2e_mcp_test.py \
31 --binary zig-out/bin/codedb \
32 --project /path/to/codedb # E2E MCP scenarios
33```
 
 
34 
35`e2e_mcp_test.py` covers three scenarios:
361. **issue-346 regression** — spawn from cwd=`/`, roots handshake, tools return real data
372. **Normal mode** — explicit positional root (`codedb <path> mcp`), immediate scan
383. **No-roots client** — spawn from `/` with no roots capability, stays alive gracefully
39 
40## Security-sensitive areas
41 
42- `src/watcher.zig` — file indexing skip lists (secrets must be excluded)
43- `src/mcp.zig` — file read/search (path traversal, scope boundaries)
44- `src/telemetry.zig` — data collection and transmission (must match docs)
45- `src/snapshot.zig` — sensitive file filtering
46- `install/install.sh` — binary download and config modification
 
 
 
 
 
 
47 
justrach/codedb · CLAUDE.md
@@ +1 @@
1# codedb — Agent Instructions
2 
3## Project
4 
5Zig 0.16.x code intelligence server. Tests live in the split `src/test_*.zig` files — one binary per area (see `build.zig`). Build and test with `zig build test`; run a single binary with e.g. `zig build test-index`.
 
 
 
 
 
 
 
6 
7## Rules
8 
9### Filing Issues
 
 
 
 
 
 
10 
11**Every GitHub issue must include a failing test case.** No exceptions.
12 
13When creating an issue:
14 
151. Write a `test "issue-XX: <description>"` block in the matching `src/test_*.zig` file (e.g. `src/test_index.zig` for index bugs) that **fails** on the current `main` branch
162. Verify it fails: `zig build test 2>&1 | grep "issue-XX"`
173. File the issue via `gh issue create` with this structure:
18 - **Title:** `<module>: <concise description>`
19 - **Body sections:** Problem, Failing Test (the zig test block), Expected, Fix
20 - **Labels:** `bug` for defects, `priority:p0` for crashes, `priority:p2` for correctness
214. Commit the failing test on a branch: `issue-XX-failing-test`
225. Do **not** fix the bug in the same commit as the failing test
23 
24If you cannot write a failing test, the issue is not well-defined enough to file.
 
 
 
25 
26### Test Style
27 
28- Use `std.testing` and `testing.allocator`
29- Use `std.heap.ArenaAllocator` for Explorer tests
30- Always `defer` cleanup (arena.deinit, allocator.free)
31- One test per issue, named `test "issue-XX: <short description>"`
32- Keep tests minimal — only exercise the specific broken code path
33 
34### Code Style
35 
36- No comments or documentation changes unless explicitly asked
37- Prefer minimal, targeted fixes over refactors
38- Follow existing patterns in the module you're editing
39 
@@ −1 +1 @@
1−# codedb Agent Guidelines
1+# codedb — Agent Instructions
22  
3−## What codedb is (and isn't)
3+## Project
44  
5−codedb is a **code-intelligence and context tool, not an editor.** It exists to
6−help agents *find and understand* code — structural search, symbol/caller
7−lookup, dependency graph, outlines, and task-shaped context — so they can edit
8−with their own native tools. codedb has **no edit capability**: the old
9−`codedb_edit` fallback (and the HTTP `POST /edit` endpoint, the post-edit
10−linter, and `codedb_diagnostics`) was removed because MCP edit calls bypass the
11−client's own undo/rewind tracking. Keep that framing consistent across tool
12−descriptions, the MCP `initialize` instructions, the README, and these docs.
5+Zig 0.16.x code intelligence server. Tests live in the split `src/test_*.zig` files — one binary per area (see `build.zig`). Build and test with `zig build test`; run a single binary with e.g. `zig build test-index`.
136  
14−## Review guidelines
7+## Rules
158  
16−- Flag any security issues: injection, file traversal, untrusted input, secret exposure
17−- Verify that sensitive files (.env, .pem, .key, credentials) are excluded from indexing AND search
18−- Check that telemetry behavior matches documentation claims
19−- Flag any regression in benchmark-critical paths (threshold: 10%)
20−- Treat P1 issues as merge-blocking
21−- Verify new language parsers handle malformed input gracefully (braces in strings, unterminated comments)
22−- Check that installer scripts don't execute untrusted code or skip verification
9+### Filing Issues
2310  
24−## Pre-merge verification
11+**Every GitHub issue must include a failing test case.** No exceptions.
2512  
26−Run these before merging any MCP-related change:
13+When creating an issue:
2714  
28−```bash
29−zig build test # unit tests
30−python3 scripts/e2e_mcp_test.py \
31− --binary zig-out/bin/codedb \
32− --project /path/to/codedb # E2E MCP scenarios
33−```
15+1. Write a `test "issue-XX: <description>"` block in the matching `src/test_*.zig` file (e.g. `src/test_index.zig` for index bugs) that **fails** on the current `main` branch
16+2. Verify it fails: `zig build test 2>&1 | grep "issue-XX"`
17+3. File the issue via `gh issue create` with this structure:
18+ - **Title:** `<module>: <concise description>`
19+ - **Body sections:** Problem, Failing Test (the zig test block), Expected, Fix
20+ - **Labels:** `bug` for defects, `priority:p0` for crashes, `priority:p2` for correctness
21+4. Commit the failing test on a branch: `issue-XX-failing-test`
22+5. Do **not** fix the bug in the same commit as the failing test
3423  
35−`e2e_mcp_test.py` covers three scenarios:
36−1. **issue-346 regression** — spawn from cwd=`/`, roots handshake, tools return real data
37−2. **Normal mode** — explicit positional root (`codedb <path> mcp`), immediate scan
38−3. **No-roots client** — spawn from `/` with no roots capability, stays alive gracefully
24+If you cannot write a failing test, the issue is not well-defined enough to file.
3925  
40−## Security-sensitive areas
26+### Test Style
4127  
42−- `src/watcher.zig` — file indexing skip lists (secrets must be excluded)
43−- `src/mcp.zig` — file read/search (path traversal, scope boundaries)
44−- `src/telemetry.zig` — data collection and transmission (must match docs)
45−- `src/snapshot.zig` — sensitive file filtering
46−- `install/install.sh` — binary download and config modification
28+- Use `std.testing` and `testing.allocator`
29+- Use `std.heap.ArenaAllocator` for Explorer tests
30+- Always `defer` cleanup (arena.deinit, allocator.free)
31+- One test per issue, named `test "issue-XX: <short description>"`
32+- Keep tests minimal — only exercise the specific broken code path
33+ 
34+### Code Style
35+ 
36+- No comments or documentation changes unless explicitly asked
37+- Prefer minimal, targeted fixes over refactors
38+- Follow existing patterns in the module you're editing
4739  
RuleStack

Built by

Kynth Studio

Directory

Configs
Stacks
Compare formats
Diff two configs
Best AGENTS.md examples

Formats

AGENTS.md
CLAUDE.md
Cursor rules
Copilot instructions

Reference

Read API
Corpus health
Privacy Policy
Terms

RuleStack

RuleStack

Built by

Kynth Studio

Directory

Configs
Stacks
Compare formats
Diff two configs
Best AGENTS.md examples

Formats

AGENTS.md
CLAUDE.md
Cursor rules
Copilot instructions

Reference

Read API
Corpus health
Privacy Policy
Terms

RuleStack

RuleStack

Built by

Kynth Studio

Directory

Configs
Stacks
Compare formats
Diff two configs
Best AGENTS.md examples

Formats

AGENTS.md
CLAUDE.md
Cursor rules
Copilot instructions

Reference

Read API
Corpus health
Privacy Policy
Terms

RuleStack