failure-forensics
Skill8 starsChanged 57 days agoMITView on GitHub
Read the wreckage before rebuilding. Use for incident post-mortems, "what happens when everything goes wrong at once?", flaky or anomalous cases being filtered out as noise, "that's weird" moments nobody wrote down, oscillating or unstable feedback behavior, and designs whose only failure mode is crash.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- voyager67/AgentFieldkittheir own version, first seen Nov 8, 2026
- soysantana/skillstheir own version, first seen Nov 1, 2026
- B-A-M-N/Confluxtheir own version, first seen Feb 25, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- soysantana/skills★ 2same name
- B-A-M-N/Conflux★ 1same name
- voyager67/AgentFieldkit★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- boundary-designSkill
- causal-auditSkill
- decision-bias-checkSkill
- estimationSkill
- evidence-synthesisSkill
- experience-and-transmissionSkill
- formal-correctnessSkill
- measurement-disciplineSkill
- narrative-sensemakingSkill
- normative-designSkill
- problem-reframingSkill
- representation-and-possibilitySkill
- structure-discoverySkill
- systems-leverageSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 42
If no shape above matches, use a built-in agent with the closest GENIUS: pattern from rules/agent-routes.json instead.
- Distinctive: line 46
No reproduction and no trace evidence exist — forensics reads records, it does not invent narratives.
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Failure Forensics
Problem shape: something failed (or will fail) and the evidence is being averaged away, cleaned up, or explained by its most convenient story. The move: treat anomalies as data, reconstruct the timeline at the right timescale, and design the degraded state as a first-class behavior.
Relevant geniuses
| Agent | Use when |
| hamilton | overload and simultaneous failure scenarios; priority shedding by criticality; the degraded mode is undesigned; "users will never do that" |
| mcclintock | aggregate metrics look smooth but one specific case is weird; a class of observations is being trimmed as noise |
| fleming | anomalies keep appearing during routine work and being cleaned up; "that's weird" said and never investigated |
| ginzburg | reconstruct what happened from marginal traces and involuntary evidence (logs, timestamps, artifacts nobody meant to leave) |
| braudel | the incident needs three-timescale decomposition — event, cycle, structure — before a root cause is named |
| wu | the failure hides under an assumption everyone considered too obvious to test |
| maxwell | the system oscillates, overshoots, or hunts — feedback stability and gain-margin diagnosis |
Invocation
- Pick the best-fit agent above. If two or more fit, run
tools/genius-invoker.sh route "<problem>"and take the top ranked match. - Load it:
tools/genius-invoker.sh invoke <agent> "<problem>", then readagents/genius/<agent>.mdin full. - Apply the agent's
<workflow>step by step and answer in its<output-format>. Reproduce before claiming a cause; classify the cause before proposing the fix (coding-standards §6 root-cause protocol). - Typical chain: ginzburg reconstructs the timeline → braudel separates
timescales → hamilton designs the degraded state. Run via
tools/genius-invoker.sh compose ginzburg hamilton -- "<problem>". - If no shape above matches, use a built-in agent with the closest
GENIUS:pattern fromrules/agent-routes.jsoninstead.
Refuse when
- No reproduction and no trace evidence exist — forensics reads records, it does not invent narratives.
- The requested output is a fix at the throw site with no cause classification.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.