agentleFS
18studio18studio/simple_okf/.agents/skills/7d/SKILL.md

7d

Skill3 starsChanged 5 months agoMITView on GitHub

Work with the repository's 7D process registry over OKF concept types. Use when validating 7D usage, generating reports by lifecycle stage, deriving feature progress, or creating/editing 7D lifecycle artifacts without changing OKF format.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. About 0 are common, 7 are distinctive, and 9 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 9

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill for 7D lifecycle work in this repository.

    line 11
  2. Very specific:

    Do not add 7D-specific frontmatter keys such as process, stage, stage_order, artifact_key, artifact_order, gate_decision, or raci solely to …

    line 19
  3. Very specific:

    Use the existing OKF type field as the lifecycle artifact type, for example PRD, Architecture & NFR, or Test Report.

    line 22
  4. Very specific:

    Do not create generic 7D types such as Lifecycle Artifact, Lifecycle Check, Role, or Decision Gate when the real artifact type is known.

    line 24
  5. Very specific:

    Check, Role, or Decision Gate` when the real artifact type is known.

    line 25
  6. Distinctive:

    Every OKF concept type must be registered.

    line 29
  7. Very specific:

    When creating or editing OKF concepts, also load and follow the okf skill.

    line 34
  8. Very specific:

    Use the simple-okf MCP server first when available:

    line 38
  9. Distinctive:

    If MCP is unavailable or the user explicitly asks for a command, use the MCP package multi-app CLI.

    line 70
  10. Distinctive:

    all seven stages in order.

    line 115
  11. Very specific:

    Do not treat missing 7D artifacts as OKF validation errors.

    line 122
  12. Distinctive:

    When the user asks to create or update 7D artifacts:

    line 128
  13. Very specific:

    Use that value as normal OKF frontmatter type and set an allowed status such as draft for a new artifact.

    line 131
  14. Distinctive:

    Place the concept in an appropriate OKF directory based on content, not on the 7D stage name alone.

    line 133
  15. Very specific:

    Run validate_7d() and validate_bundle().

    line 139
  16. Distinctive:

    When reporting 7D work, summarize:

    line 143

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

7D Skill

Use this skill for 7D lifecycle work in this repository.

7D is a process registry over existing OKF concept types. It does not change OKF concept format. The canonical rules are in SPEC.md, AGENTS.md, and the 7D registry implemented by okf_mcp/okf.py.

Core rules

  1. Do not add 7D-specific frontmatter keys such as process, stage, stage_order, artifact_key, artifact_order, gate_decision, or raci solely to support 7D.
  2. Use the existing OKF type field as the lifecycle artifact type, for example PRD, Architecture & NFR, or Test Report.
  3. Do not create generic 7D types such as Lifecycle Artifact, Lifecycle Check, Role, or Decision Gate when the real artifact type is known.
  4. Resolve stage and RACI responsibility from the registry, not from per-concept metadata.
  5. Derive feature progress from linked artifact concepts and their types.
  6. Every OKF concept type must be registered. An unmapped type is a validation error; missing required lifecycle artifacts are static-analysis coverage gaps.
  7. OKF concepts also require status. Allowed values are draft, to-review, not-valid, valid, rejected, and accepted; status is document/artifact state, not implementation progress, and transitions are recommended only.
  8. When creating or editing OKF concepts, also load and follow the okf skill.

Preferred MCP tools

Use the simple-okf MCP server first when available:

seven_d_registry()
seven_d_mapping_for_type(type_name)
list_7d_artifact_concepts(stage?)
seven_d_stage_report(stage?)
seven_d_stage_report_markdown(stage?)
seven_d_dashboard(write?, out_path?, include_html?)
seven_d_feature_status(concept_id)
validate_7d()
validate_bundle()
build_graph(write=false)

Typical workflows:

  • Registry lookup: call seven_d_registry() or seven_d_mapping_for_type(type_name).
  • Stage report: call seven_d_stage_report(stage) for structured data, or seven_d_stage_report_markdown(stage) for a compact Markdown report.
  • Kanban dashboard: call seven_d_dashboard(write=true, out_path="artifacts/7d-dashboard.html") to generate a self-contained HTML board with stage columns and readonly concept detail modals.
  • Stage inventory: call list_7d_artifact_concepts(stage) for one stage, or for every stage in the registry.
  • Feature progress: call seven_d_feature_status(concept_id) and report the derived stage, supporting artifacts, and gaps.
  • Validation: call validate_7d() and validate_bundle() before finishing.

CLI helper tools

If MCP is unavailable or the user explicitly asks for a command, use the MCP package multi-app CLI. Resolve paths from the repository root.

Generate a compact report for every 7D stage:

python3 -m okf_mcp 7d report --bundle okf
python3 -m okf_mcp 7d report --bundle okf --stage Design

Generate the interactive Kanban dashboard HTML:

python3 -m okf_mcp 7d dashboard --bundle okf --out artifacts/7d-dashboard.html

Return the same report as JSON:

python3 -m okf_mcp 7d report --bundle okf --json

Validate 7D usage:

python3 -m okf_mcp 7d validate --bundle okf

Show one feature's derived 7D status:

python3 -m okf_mcp 7d status --bundle okf requirements/flows/first-reasoning-onboarding

Show the registry:

python3 -m okf_mcp 7d registry --bundle okf

Stage report expectations

A stage report should include:

  1. bundle path and validation status;
  2. all seven stages in order;
  3. registered artifact types for each stage;
  4. actual OKF concepts found for each stage;
  5. explicit static-analysis gaps for missing required lifecycle artifacts;
  6. unmapped type validation errors, if any;
  7. warnings/errors from validate_7d().

Do not treat missing 7D artifacts as OKF validation errors. They are lifecycle coverage gaps unless the user defines a stricter gate. Do treat unmapped concept type values as validation errors.

Creating lifecycle artifacts

When the user asks to create or update 7D artifacts:

  1. Identify the correct registered artifact type.
  2. Use that value as normal OKF frontmatter type and set an allowed status such as draft for a new artifact.
  3. Place the concept in an appropriate OKF directory based on content, not on the 7D stage name alone.
  4. Link the artifact to the feature or related concepts using relative Markdown links.
  5. Regenerate indexes/graph only when concepts are added, removed, renamed, or links materially change.
  6. Run validate_7d() and validate_bundle().

Output format

When reporting 7D work, summarize:

  1. stage coverage and derived feature status;
  2. registered artifact types used;
  3. gaps or unknown types;
  4. validation results;
  5. files changed, if any.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.