Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/microsoft/tscodesearch/agents-mdgit clone --depth 1 https://github.com/microsoft/tscodesearchWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.06931 | $0.06931 |
| Opus 5 | $0.03465 | $0.03465 |
| Sonnet 5 | $0.01386 | $0.01386 |
| Haiku 4.5 | $0.00693 | $0.00693 |
Grade A, and why
tscodesearch AGENTS.md scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Management API via curl (read key/port from config.json -- never hard-code): How it starts
The opening of the file, as written. The whole thing — 418 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent notes
This is the canonical instructions file for repository guidance. Put all long-form agent instructions in AGENTS.md. CLAUDE.md is a placeholder that points here.
Operational notes for agents working in this repo are maintained below.
GitHub workflow preference
This repository is hosted on GitHub. When working with pull requests, comments, checks, issues, and repository metadata, prefer GitHub-native tooling (for example, GH CLI and available GitHub-integrated tools) whenever possible.
Debugging the indexer: real-time CSV logs
When something goes wrong on the indexer side (the classic symptom: files get re-indexed on every restart even though the index already contains them), turn on the CSV debug log.
Enable
Set csv_debug in config.json:
{
"port": 8108,
"csv_debug": true,
"roots": { "default": { "path": "q:/spocore/src" } }
}
Then node ts.mjs restart (or ts restart if the cmd is on PATH).
Accepted values:
| value | effect |
|---|---|
false / "" / unset |
off |
true / "1" / "on" / "true" |
logs to <repo>/.tantivy_csv/ |
| any other string | treated as an explicit directory path -- logs land there |
Logging is opt-in and append-only; rows from successive restarts share a
file and are distinguished by the per-row pid column.
CSV files
Written by indexserver/debug_logger.py, one file per event type. Each row
starts with ts (ms-precision local time) and pid.
| file | rows are written when | useful columns |
|---|---|---|
session.csv |
daemon start/stop | action = start/stop |
backend_export.csv |
verifier exports the index map at scan start (once per session) | doc_id, mtime, relative_path |
fs_walk.csv |
every file the verifier walks against the exported map | decision = matched / missing / stale |
orphan.csv |
index entries with no matching file on disk (about to be deleted) | doc_id |
enqueue.csv |
every IndexQueue.enqueue (including dedup hits) |
action, reason, is_new |
parse.csv |
every tree-sitter parse the queue worker runs | parse_ms, ok, error |
commit.csv |
every Tantivy commit, success or failure | duration_ms, success, error |
watcher.csv |
every watchdog event | src_path, action |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 418 lines · 6,931 tokens per session scan A e6c3e9fcf9bd
tscodesearch AGENTS.md is an instructions file published in the GitHub repository microsoft/tscodesearch (4 stars, last pushed 8d ago), licensed MIT. It adds 6,931 tokens to every session, about $0.0347 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.