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 skills add atuljha23/holocron --skill observability-checklistgit clone --depth 1 https://github.com/atuljha23/holocronWrote 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/atuljha23/holocron/observability-checklist)<a href="https://agentmods.dev/skills/atuljha23/holocron/observability-checklist"><img src="https://agentmods.dev/badge/skills/atuljha23/holocron/observability-checklist/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/atuljha23/holocron/observability-checklist"><img src="https://agentmods.dev/badge/skills/atuljha23/holocron/observability-checklist.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 | $0.00049 | $0.01067 |
| Opus 5 | $0.00024 | $0.00534 |
| Sonnet 5 | $0.00010 | $0.00213 |
| Haiku 4.5 | $0.00005 | $0.00107 |
Grade A, and why
observability-checklist 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 12d 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.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Observability checklist
The goal isn't logs — it's answering "what happened" in under five minutes at 2am. Design backward from the question.
Logs
Structured or nothing
- JSON (or logfmt), not free text.
- Every log has:
ts,level,service,request_id,msg, and relevant fields. - Fields are snake_case or camelCase — pick one per service. No mixing.
- Durations in milliseconds as numbers, not "took 2s" strings.
Levels are a contract
error: something failed the user can't recover from. Alert-worthy.warn: something unexpected but handled. Review-worthy.info: the shape of normal traffic. Sampling OK.debug: developer-only. Off in prod.- Don't invent new levels.
What to log
- Request entry (method, path, user, request_id).
- Outcome (status, duration).
- Branch decisions that matter (authz check, cache hit/miss, retry).
- Errors with the full context you need to reproduce.
What NOT to log
- Secrets, tokens, passwords, PII — scrub before logging.
- The full request body by default. Specific fields, yes. Everything, no.
- Stack traces as separate lines — include as a
stackfield so they stay with the event.
Traces
Span per unit of work
- HTTP request handler = 1 span.
- Outbound call = child span.
- DB query = child span (with
db.statementtruncated / hashed, not raw if PII risk). - Background job step = 1 span.
Mandatory attributes
service.name,service.version.- For HTTP:
http.method,http.route,http.status_code. - For DB:
db.system,db.operation. - For errors:
exception.type,exception.message.
Propagation
- Every outbound request carries trace context headers.
- Every background job enqueues with trace context.
- A trace that ends at the HTTP edge and restarts at the worker is useless.
Metrics
RED for services
- Rate — requests per second, per endpoint.
- Errors — error rate, per endpoint.
- Duration — p50/p95/p99, per endpoint.
USE for resources
- Utilization — % of capacity in use.
- Saturation — queue depth / wait time.
- Errors — failed operations.
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.
- 12d ago First seen · 114 lines · 49 tokens per session scan A 0a4f8bd940b1
observability-checklist is a skill published in the GitHub repository atuljha23/holocron (2 stars, last pushed 4mo ago), licensed MIT. It adds 49 tokens to every session and 1,067 once invoked, about $0.0002 per session on Opus 5. 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 skills, from other repositories
bug-capture
Capture a user-reported defect as a durable GitHub issue written in the project's own domain language. Explores the codebase in parallel for context but never leaks file paths or line numbers into the issue. Use when the user reports a bug conversationally, runs a QA pass, or says "file an issue", "log this as a bug"…
debugging-testing
Use when the user reports a bug they can't reproduce, asks where to start debugging, or mentions a Heisenbug / production-only failure. Drives the observe→hypothesize→predict→test→iterate scientific method.
gsd-preflight
Use when GSD commands are failing or .planning/ artifacts look inconsistent. Run before any GSD execute-phase or verify-work attempt to catch cross-artifact mismatches early.
ap-depth-prober
Report the compatibility redirect to ap-worker; this retired role cannot perform new work.
ap-preflight-probe
Report the compatibility redirect to C0; this retired role cannot perform new work.
oversight-team
Use when you are a seat on the oversight pod (a standing monitor-mode rig that keeps OTHER rigs healthy), configuring or running the drift detectors, or choosing whether to intervene vs escalate. Covers the pull-not-poll posture, the v0 detectors (premature-park, process-drift, off-task, token-burn), the intervention…