agentsclimarketplace
Sign in

Project control

Skill hunterbohm/skills/skills/project-management/project-control

Project lens that builds an adaptive Codex App visualization from the active project's real sources. Use for project orientation, planning, verification, scenario exploration, or refresh.From its SKILL.md

Install
npx -y skills add hunterbohm/skills --skill project-control

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

  • 0 stars0 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

2.8 KB, 541 tokens by cl100k_base, as published. Nobody here has run it

Project Control

Create a current project lens inside the Codex App: a disposable visual over canonical project sources, never a second tracker.

Compatibility

Codex App only. This release requires Codex inline visualizations, the model-invoked visualize skill, and window.openai.sendFollowUpMessage. Other agent harnesses are not supported yet.

Run

  1. Route. Read the nearest AGENTS.md or CONTEXT.md and resolve one owning workspace. Complete when one project and its source-of-truth surfaces are named.
  2. Read. Inspect the smallest useful set of canonical sources: overview, tracker or task files, recent Git state, proof, decisions, and release material. Complete when every important claim in the proposed view has a source, uncertainty label, or explicit user report.
  3. Lens. Answer the universal project questions in references/project-lens.md without forcing project-specific concepts into fixed fields. Complete when the current situation, implication, blockers or unknowns, and next moves are clear in plain language.
  4. Choose. Select the smallest useful visual from references/view-selection.md. Complete when each chosen view exposes a distinct, source-supported relationship the user needs to understand.
  5. Render. Load Codex's model-invoked visualize skill, then build the in-task experience with the project-specific rules in references/codex-app.md. Complete when the first render is useful, interactions work, and the layout passes wide and narrow inspection without errors.
  6. Explain. Lead with what is happening and why it matters, then show the visual. Complete when the response stands alone in plain language.

Branches

  • Orient: current reality, blockers, dependencies, decisions, and next moves.
  • Plan: sequence, milestones, ownership, and critical path.
  • Verify: criteria, tests, claims, receipts, and release gates.
  • Explore: a user-named variable or tradeoff, clearly labeled hypothetical.
  • Refresh: re-read canonical sources, regenerate the lens, and call out material changes.

Guardrails

  • Facts, inferences, and scenarios remain visibly distinct; missing evidence stays unknown.
  • Metrics appear only when sourced or transparently derived.
  • Project language stays plain; technical terms receive an ordinary-language explanation.
  • The default run is read-only. Permanent boards, tracker edits, code changes, and external writes require an explicit request.
  • Ambiguous ownership triggers one concise routing question.

What ships with it: 5 files

6.9 KB alongside SKILL.md

agents/

references/

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.