Generate regression cases

Skill xiaolai/eou-foundry/skills/generate-regression-cases

Convert incident reports and audit failures into candidate regression cases that prevent re-occurrence — each case captures a concrete observable symptom, the target EOU, the failure class (F-code), and an expected-behavior predicate. Cases are written as candidates; activation requires human owner approval. <example> Context: A diagnosis was filed for a path-drift incident; owner wants regression coverage so the failure cannot return silently. user: "$generate-regression-cases foundry/incidents/inc-0042.yml" assistant: "I'll extract the observable symptom, classify it under the F-code taxonomy, and write a regression case under foundry/self-evolution/regression/cases/. activation_status defaults to candidate." </example> <example> Context: User wants regression cases for a batch of audit findings. user: "$generate-regression-cases foundry/audits/eou-audits/*.audit.yml" assistant: "I'll iterate each finding, generate at most one regression case per observable symptom, and refuse to generate cases for findings that name no observable symptom (a label without a symptom is not accepted)." </example>From its SKILL.md

Install

npx -y skills add xiaolai/eou-foundry --skill generate-regression-cases

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

9 stars9 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.0 KB, as published. Nobody here has run it

Generate Regression Cases

Convert $source (one or more incident reports or audit failures) into candidate regression cases.

Inputs

$source (required) — path to one or more incident reports or audit failure files. Accepts a direct file path or a glob (e.g., foundry/audits/incidents/*.yml). Every matched file must contain at least one observable failure symptom.

Required reading

foundry/failure-taxonomy.yml — classify each failure by F-code
foundry/constitution.yml — invariants that constrain what counts as a valid regression case
schemas/incident.schema.yml — validate structured YAML incident files against this schema
schemas/regression-case.schema.yml — the schema for the output regression case files

Stop conditions

Halt and report before generating cases if:

$source resolves to no readable files.
No observable failure symptom can be extracted from any source file — labels and conclusions without at least one observable symptom (error message, wrong output, missing artifact, schema mismatch) are not accepted as source input.
schemas/regression-case.schema.yml does not exist — cannot validate output structure.

Procedure

Load $source. If it is a glob, read all matching files.
For each incident or audit failure, extract the observable failure symptom.
Classify the failure using foundry/failure-taxonomy.yml (F1–F12).
Derive the regression case: the minimal input that reproducibly triggers the failure.
Construct the fixture path: foundry/self-evolution/regression/fixtures/{case_id}/.
Check foundry/self-evolution/regression/cases/ for existing cases that cover this failure — skip if a duplicate exists; record the duplicate case ID instead.
For each new case, write the regression case file with all required fields.
Set activation_status: candidate on all generated cases — never active.

Required fields per case (per `schemas/regression-case.schema.yml`)

id:                   # regression-{eou_id}-{failure_class}-{YYYYMMDD}
source_incident_id:   # incident or audit ID this case was derived from
failure_class:        # F1-F12 code
eou_id:               # EOU that exhibited the failure
fixture_path:         # foundry/self-evolution/regression/fixtures/{case_id}/
failure_shape_assertion:  # observable output that indicates the failure is occurring (non-empty)
activation_status:    # candidate — never active without human approval

Output

Write each case to foundry/self-evolution/regression/cases/{case_id}.regression.yml. Create the fixture directory at foundry/self-evolution/regression/fixtures/{case_id}/. Record the run in foundry/runs/{eou_id}/{run_id}.yml.

Constraints

Do not set activation_status: active. Cases are candidates until the human owner promotes them.
Do not generate cases without a concrete source incident or audit failure.
Each case must include at least one failure_shape_assertion — what observable output signals the failure is occurring. Cases asserting only pass are invalid.
If a duplicate case exists for this failure, record the duplicate ID and skip generation.

Scope Note

Upstream: receives incidents (foundry/incidents/) or audit findings (foundry/audits/eou-audits/).

Downstream: produces regression cases at foundry/self-evolution/regression/cases/ at activation_status candidate. Cases enter the regression suite only after human owner approval — this skill never sets activation_status to active.

Related: $eou-diagnose (sibling — F-code classification of the same incident); $ecp-propose (sibling — proposes change; regression cases pin the failure so the change can be verified).

Pipeline: incident | audit failure → generate-regression-cases → human approval → activation_status: active