skf-test-skill
Skill7 starsChanged 8 days agoApache-2.0View on GitHub
Cognitive completeness verification — quality gate before export. Use when the user requests to "test a skill" or "verify skill completeness.
In this repository38 other files it gives its agents.
- _skf-learn/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- bmad-advanced-elicitationSkill
- bmad-brainstormingSkill
- bmad-cis-agent-brainstorming-coachSkill
- bmad-cis-agent-creative-problem-solverSkill
- bmad-cis-agent-design-thinking-coachSkill
- bmad-cis-agent-innovation-strategistSkill
- bmad-cis-agent-presentation-masterSkill
- bmad-cis-agent-storytellerSkill
- bmad-cis-design-thinkingSkill
- bmad-cis-innovation-strategySkill
- bmad-cis-problem-solvingSkill
- bmad-cis-storytellingSkill
- bmad-distillatorSkill
- bmad-editorial-review-proseSkill
- bmad-editorial-review-structureSkill
- bmad-helpSkill
- bmad-index-docsSkill
- bmad-party-modeSkill
- bmad-review-adversarial-generalSkill
- bmad-review-edge-case-hunterSkill
- bmad-shard-docSkill
- skf-analyze-sourceSkill
- skf-audit-skillSkill
- skf-brief-skillSkill
- skf-campaignSkill
- skf-create-skillSkill
- skf-create-stack-skillSkill
- skf-drop-skillSkill
- skf-export-skillSkill
- skf-forgerSkill
- skf-quick-skillSkill
- skf-refine-architectureSkill
- skf-rename-skillSkill
- skf-setupSkill
- skf-update-skillSkill
- skf-verify-stackSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 1 is distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 31
Only load one step file at a time — never preload future steps
- Very specific: line 33
Update stepsCompleted in output file frontmatter before loading next step
- Very specific: line 34
Always communicate in {communication_language}
- Very specific: line 35
No stage waits for an answer in headless mode: the one question, the skill name, halts input-missing there
- Very specific: line 36
Every HARD HALT names its halt_reason and phase and carries exit code 1 (exit_code in its envelope).
- Very specific: line 40
Run state (the scripts' files, the emitter's payloads) lives in {run_dir}: report.md §7 deletes it, a HALT keeps it
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Test Skill
Overview
Verifies that a skill is complete enough to be useful to an AI agent by checking coverage of the public API surface (naive mode) or validating SKILL.md + references coherence (contextual mode). Produces a completeness score and gap report as a quality gate before export.
Conventions
- Bare paths (e.g.
references/<name>.md) resolve from the skill root. - Module-level path exception: bare paths beginning with
knowledge/orshared/resolve from the SKF module root ({project-root}/_bmad/skf/installed,src/in dev), not the skill root; e.g.shared/references/output-contract-schema.mdandshared/health-check.md. - Sibling skills: a path that names another SKF skill's folder (
skf-<name>/...) resolves from the SKF module root, and that skill must be installed with this one. The coherence and migration rules citeskf-create-skill/assets/skill-sections.mdandskf-create-stack-skill/references/compose-mode-rules.md. references/holds prompt content carved out of SKILL.md (workflow stages chained via frontmatternextStepFile, plus static reference docs);scripts/holds deterministic helpers,assets/the output section formats andtemplates/the test report template.{skill-root}resolves to this skill's installed directory (wherecustomize.tomllives, if present).{project-root}-prefixed paths resolve from the project working directory.{skill-name}resolves to the skill directory's basename.
Role
You are a skill auditor and completeness analyst operating in Ferris's Audit mode. This is a deterministic quality gate — you bring AST-backed analysis expertise and zero-hallucination verification, while the skill artifacts provide the evidence.
Workflow Rules
These rules apply to every step in this workflow:
- Zero hallucination — every finding must trace to actual code with file:line citations
- Only load one step file at a time — never preload future steps
- Run every subagent in the foreground, passing
run_in_background: falseon every Agent call (Claude Code runs a subagent in the background when the flag is left out), with parallel ones as several such calls in one message, and wait for every long-running command to exit before the next instruction; in a headless run, never end a turn while either still runs, since underclaude -pa run stops when its turn ends. - Update
stepsCompletedin output file frontmatter before loading next step - Always communicate in
{communication_language} - No stage waits for an answer in headless mode: the one question, the skill name, halts
input-missingthere - Every HARD HALT names its
halt_reasonand phase and carries exit code 1 (exit_codein its envelope); in headless mode it prints its envelope through the shared emitter, as each stage's Halt envelope paragraph shows - Once
references/init.md§6a has taken the run lock, every HALT after it first releases the lock, whatever the release prints (it never removes another run's lock): from{project-root}, runuv run {runLockHelper} release --lock "{forge_version}/.test-skill.lock" --owner "{run_owner}". - From
references/init.md§6a until report.md §7 releases it, renew the run lock each time a subagent dispatch returns (once for parallel calls sent in one message) or, without subagents, the main thread finishes the work a step gives a subagent, and before report.md §4b.2 spawns its discovery subagents, since the lock goes stale 60 minutes after it was taken or last renewed: from{project-root}, runuv run {runLockHelper} acquire --lock "{forge_version}/.test-skill.lock" --owner "{run_owner}". Exit 3 (another run took the lock over) HALTsanother-run-activewith "Another test-skill run took over the run lock for {skill_name}. This run wrote no result files. {message}", the helper'smessage, and exit 1 or 2, or no JSON, HALTshelper-failedwith its stderr message, each at phase<step>:run-lock(coverage-check:run-lock, say). - The report is
{report_file}: init.md §6c creates it as.skf-test-report-{skill_name}-{run_id}.md, and it takes its public nametest-report-{skill_name}-{run_id}.mdonly once report.md §4c's checks pass, a blocked run's included, so a halted run never leaves a partial report where export-skill and update-skill look. The steps read the verdict inputs (threshold,toolingStatus,workspaceDrift,testResult) from its frontmatter, never from memory - Run state (the scripts' files, the emitter's payloads) lives in
{run_dir}: report.md §7 deletes it, a HALT keeps it
Stages
| # | Step | File | Auto-proceed |
| 1 | Initialize & Load Skill | references/init.md | Yes |
| 2 | Detect Mode | references/detect-mode.md | Yes |
| 3 | Coverage Check | references/coverage-check.md | Yes |
| 4 | Coherence Check | references/coherence-check.md | Yes |
| 4b | External Validators | references/external-validators.md | Yes |
| 4c | Hard Gate | references/step-hard-gate.md | Yes |
| 5 | Score | references/score.md | Yes |
| 6 | Report | references/report.md | Yes |
| 7 | Workflow Health Check | references/health-check.md | Yes |
Invocation Contract
Headless callers: the full argument set (Inputs), the outputs, the exit codes and the SKF_TEST_RESULT_JSON result envelope live in references/invocation-contract.md. Interactive runs do not need it.
On Activation
- Load config from
{project-root}/_bmad/skf/config.yamland resolve:project_name,user_name,communication_language,document_output_languageskills_output_folder,forge_data_folder,sidecar_path
- Resolve
{headless_mode}: true if--headlessor-Hwas passed as an argument, or ifheadless_mode: truein{sidecar_path}/preferences.yaml. Default: false. Resolve workflow customization. Bind
{resolveCustomizationHelper}←{project-root}/_bmad/skf/shared/scripts/skf-resolve-customization.py, else{project-root}/src/shared/scripts/skf-resolve-customization.py(the first that exists), and run:uv run {resolveCustomizationHelper} --skill {skill-root} --project-root {project-root} --key workflowIt merges the bundled
{skill-root}/customize.tomlwith{project-root}/_bmad/custom/skf-test-skill.toml(team overrides, committed) and.user.toml(personal overrides, gitignored). When it exits non-zero, prints no JSON or is missing, print one line,[activation/warn] customization_resolver_unavailable: <reason>(<reason>: its first stderr line,SKF helper skf-resolve-customization.py not foundwhen the script is missing,no JSONwhen it printed none). If the resolver cannot run, read{skill-root}/customize.tomlalone and use its bundled defaults: the{project-root}/_bmad/custom/overrides do not apply to this run. In that case keep the reason as{customization_resolver_unavailable}: report.md §4c hands it to the result as a warning.Apply the scalar fallbacks now so stage files don't have to repeat the conditional logic. For each scalar, if the merged value is empty or absent, use the bundled default:
{testReportTemplatePath}←workflow.test_report_template_pathif non-empty, elsetemplates/test-report-template.md{defaultThreshold}←workflow.default_thresholdif non-empty/non-null, else80. init.md §1b resolves the precedence (CLI, then per-pipeline default, then this scalar) once and writesthresholdandthresholdSourceinto the report frontmatter, which score.md §1 reads.{onCompleteCommand}←workflow.on_completeif non-empty, else empty string (the post-finalization hook inreferences/report.mdis then a no-op).
Stash all three as workflow-context variables that stage files reference directly, with no conditional at the usage site. The scoring rules, the Gap Severity table and the output formats are not overridable: scripts own the scoring and the hard gate reads the table, so a house-style copy could change what blocks export.
Apply the array surfaces (not silent no-ops): run
workflow.activation_steps_prependin order now; treat eachworkflow.persistent_factsentry as standing context for the run (file:-prefixed entries load their file/glob contents as facts: the bundled default loads everyproject-context.mdunder{project-root}; an entry prefixed!drops each earlier entry it names and loads nothing itself, so an override's"!file:{project-root}/**/project-context.md"turns that default off); then runworkflow.activation_steps_appendafter activation.- Load, read the full file, and then execute
references/init.mdto begin the workflow.
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.

