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 commands/bharath03-a/gitcortex/token-benchmarkgit clone --depth 1 https://github.com/bharath03-a/GitCortexWhat 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 | $0.00000 | $0.00377 |
| Opus 5 | $0.00000 | $0.00188 |
| Sonnet 5 | $0.00000 | $0.00075 |
| Haiku 4.5 | $0.00000 | $0.00038 |
Grade A, and why
token-benchmark scanned grade A with 0 findings 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 2d ago.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run the REAL token benchmark (measured Claude usage) and turn it into a ship/fix decision. Use before a release or after changing any MCP tool.
Delegate to the token-benchmark agent. Pass along any arguments in $ARGUMENTS:
- a model id (default
claude-haiku-4-5-20251001), - a scope hint (
full= 5 languages, or a single repo/tool name), diffto compare against the previous run.
The agent will:
- Build
target/release/gcxif stale. - State the cost estimate (full haiku sweep ≈ $5–8, 20–40 min; each session
capped by
BUDGET, default $0.75) before launching anything. - Run
docs/benchmarks/real-sweep.sh(or a singlereal-harness.shfor a quick check), which runs Claude twice per question — grep arm vs GitCortex MCP arm — and records realusagetokens, cost, and turns. - Render the tool × language scorecard:
python3 docs/benchmarks/real-report.py→docs/benchmarks/real-report.html. - Diff against the previous run; flag any tool that went green→red as a regression.
- Report a per-tool verdict (keep / marginal / REDESIGN), a concrete minimal fix for each REDESIGN tool, and a one-line ship-or-block call.
Do not let it edit MCP code — it proposes fixes; apply them via the
add-mcp-tool skill, then re-run this to confirm the fix worked.
See docs/benchmarks/RELEASE-GATE.md for the full method.
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.
- 2d ago First seen · 27 lines · 0 tokens per session scan A e1fbbd9c9141
token-benchmark is a command published in the GitHub repository bharath03-a/GitCortex (7 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 377 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.