Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/odere-pro/claude-calibrationnpx agentmods add skills/odere-pro/claude-calibration/calibration-doctorWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/odere-pro/claude-calibration/calibration-doctor)<a href="https://agentmods.dev/skills/odere-pro/claude-calibration/calibration-doctor"><img src="https://agentmods.dev/badge/skills/odere-pro/claude-calibration/calibration-doctor/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/odere-pro/claude-calibration/calibration-doctor"><img src="https://agentmods.dev/badge/skills/odere-pro/claude-calibration/calibration-doctor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00137 | $0.01248 |
| Opus 5 | $0.00068 | $0.00624 |
| Sonnet 5 | $0.00027 | $0.00250 |
| Haiku 4.5 | $0.00014 | $0.00125 |
Grade B, and why
calibration-doctor scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
a triage list. Do not run any scripts yourself; do not write any files. How it starts
The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
echo "=== calibration-doctor preprocessing ==="
PROJECT_DIR="${CLAUDE_PROJECT_DIR:-$(pwd)}"
SCRIPT="${CLAUDE_SKILL_DIR}/scripts/doctor.sh"
echo "PROJECT_DIR=$PROJECT_DIR"
echo "DOCTOR_SCRIPT=$SCRIPT"
if [ -x "$SCRIPT" ] || [ -f "$SCRIPT" ]; then
echo "--- doctor.sh output (TSV) ---"
bash "$SCRIPT" "$PROJECT_DIR" 2>&1
echo "--- end doctor.sh output ---"
else
echo "DOCTOR_SCRIPT_NOT_FOUND=$SCRIPT"
fi
echo "=== end preprocessing ==="
/claude-calibration:calibration-doctor
You are the calibration doctor. The preprocessing block above has already run
scripts/doctor.sh "$PROJECT_DIR" and inlined its TSV output between
--- doctor.sh output (TSV) --- markers. Your job is to format and print that output as
a triage list. Do not run any scripts yourself; do not write any files.
The TSV is one row per check: <check>\t<status>\t<detail>, where status is one of:
ok— check passed.warn— non-blocking concern (style threshold, optional setup).broken— config likely doesn't work (missing file, parse error, missing required field).
Optional argument: $ARGUMENTS — if it contains the token --verbose, include the ok
rows in the output. Otherwise hide them and only show the counts.
Output format
Always start with a one-line header:
calibration-doctor · <PROJECT_DIR>
Then group findings by status, broken first, then warn. Each row is
<status-glyph> <check> — <detail>:
✗ broken: agent:foo.md — missing required frontmatter: tools
✗ broken: hook:format.sh — script not executable (chmod +x): ./hooks/format.sh
! warn: claude-md:size — 837 lines (>800 — split or trim; see calibrate-claude-md)
! warn: mcp:linear — command not on PATH: linear-mcp (server may not start)
If $ARGUMENTS contains --verbose, also list the ok rows under a Healthy: heading.
Close with a count line and a → Next: pointer:
<B> broken · <W> warn · <O> ok
→ <pointer>
Where the pointer depends on what fired:
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 119 lines · 137 tokens per session scan B 91514c9ea377
calibration-doctor is a skill published in the GitHub repository odere-pro/claude-calibration (1 stars, last pushed 1mo ago), licensed MIT. It adds 137 tokens to every session and 1,248 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it B with 1 finding (unrestricted tool access). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
verifying-external-behavior
Confirms what a third-party library, remote API, build backend, or scraped document actually does before writing code that depends on it — throwaway probes that run in seconds, permissive clients that forward wrong arguments instead of rejecting them, per-endpoint docs that don't generalize, response shapes that make…
codebase-navigator
Deep structural codebase comprehension, AST dependency graph traversal, symbol caller/callee tracing, and monorepo navigation strategies for massive codebases.
root-cause-analysis
Hyper-scale root cause analysis (RCA), bisection, call-tree backtracking, and automated surgical remediation across massive million-file enterprise repositories.
kaboom-connection-guard
Use when a tool call fails with "extension not connected", the daemon is unreachable, or browser telemetry stops arriving.
debug-triage
Use when user reports a broken page, unexpected behavior, or error and needs fast root-cause identification across multiple signal sources.
debug
Use when user reports a bug, broken feature, console error, failed network request, or unexpected page behavior that needs investigation.