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 3xachris/3xa-harness --skill staged-diagnosisgit clone --depth 1 https://github.com/3xachris/3xa-harnessWrote 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/3xachris/3xa-harness/staged-diagnosis)<a href="https://agentmods.dev/skills/3xachris/3xa-harness/staged-diagnosis"><img src="https://agentmods.dev/badge/skills/3xachris/3xa-harness/staged-diagnosis/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/3xachris/3xa-harness/staged-diagnosis"><img src="https://agentmods.dev/badge/skills/3xachris/3xa-harness/staged-diagnosis.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 | $0.00079 | $0.00877 |
| Opus 5 | $0.00039 | $0.00439 |
| Sonnet 5 | $0.00016 | $0.00175 |
| Haiku 4.5 | $0.00008 | $0.00088 |
Grade A, and why
staged-diagnosis 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 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.
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 — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Staged Diagnosis
The six stages are ordered because each one produces the evidence the next one runs on, and a stage is finished when it has produced that thing. Each edit made before its stage has produced evidence is one the next investigation has to account for.
The six stages
- Reproduce — get to one command you can re-run that goes red on this failure: the trigger, its input, its environment flags, its machine state. Produces: that command, and the red result it gives. A path you can walk by hand is the weaker version of this and only acceptable while the automatic one is out of reach — everything downstream is measured by whether this goes green, so a loop you have to drive by hand slows every later stage. Where it fires only sometimes, run it repeatedly and record what varies; the shape of the intermittency is the strongest clue on offer.
- Minimise — cut away everything the failure survives without, until it sits in one layer: the script, the service, the data, the encoding, the environment. Produces: the smallest case that still fails, and the layer it lives in.
- Hypothesise — write the causal claim down in one sentence, in terms of that layer. Produces: a statement specific enough to be wrong. Check the project's own record of past failures first — a familiar shape is a precedent to apply, not a discovery to repeat.
- Instrument — put the hypothesis in front of evidence: a log line, a printed value, a breakpoint, a checksum. Produces: a measurement that confirms or kills the claim. Edits belong after this stage, so that what gets changed is what the evidence pointed at.
- Fix — change the confirmed cause and nothing beside it. Produces: one focused edit. When the fix would reach into frozen specification, another task's settled rules, or a shared module several callers depend on, stop and ask — the cheapest version of that conversation happens before the edit.
- Regression-test — re-run the original trigger, then the checks around it. Produces: evidence the failure is gone and its neighbours still work, attached to the closeout.
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 · 34 lines · 79 tokens per session scan A b75790c9014c
staged-diagnosis is a skill published in the GitHub repository 3xachris/3xa-harness (17 stars, last pushed 11d ago), licensed MIT. It adds 79 tokens to every session and 877 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
hotfix
Fixes an observed defect with reproducible evidence in one call: writes a short trace doc before touching code, implements the fix, and backs it with a regression test written before the fix. Production incidents are the motivating case, not a gate. When blocked, it halts by name and saves the doc for a later call to…
testing
Characterization testing and safety-net backfill for existing code. Use when legacy, under-tested, or risky code needs tests before a refactor, bug fix, or behavior change. Captures current behavior through public interfaces, identifies coverage gaps, and adds focused unit, integration, or E2E tests without replacing…
autonomous-tdd-debugger
Empowers the agent to autonomously run tests, read terminal stack traces, and self-heal code until tests pass. Transforms the agent from a passive coder to an active CI pipeline debugger.
debug-flow
Systematic debugging workflow. Reproduce the issue, isolate root cause, write a failing test, fix, verify. Use when diagnosing bugs or unexpected behavior.
debugging
Investigate failures whose root cause is still unknown — narrow the search space, instrument, and test falsifiable hypotheses. Use for intermittent or environment-dependent behavior, unexplained stack traces, regressions with no known trigger, or any symptom without a confirmed cause. Ends once the root cause is…
testing
Strategy and rules for writing or improving automated tests. Use when adding tests, improving coverage, fixing flaky tests, setting up a test suite, or deciding what and how to test. Emphasizes deterministic tests, testing behavior over implementation, and a tight validation loop. For diagnosing production failures…