CtrlK
BlogDocsLog inGet started
Tessl Logo

pantheon-ai/implementation-planner

Converts a PRD or requirements document into a structured, phased implementation plan with individual phase files and granular per-task files written to .context/plans/. Also restructures existing monolithic planning documents into digestible, hierarchical directory structures. Creates a root plan index summarising all phases, a numbered phase file per phase, and a numbered task file per task inside each phase directory.

92

3.25x
Quality

93%

Does it follow best practices?

Impact

91%

3.25x

Average score across 5 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-5/

{
  "context": "Agent was asked to add a new phase (REST API layer) to an existing plan at .context/plans/plan-blog-platform/. Evaluate the new phase-03 files only — existing phase-01 and phase-02 must remain unchanged.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "phase-03-directory-created",
      "description": "A new directory phases/phase-03-<slug>/ was created inside the existing plan (not a new plan root)",
      "max_score": 6
    },
    {
      "name": "existing-phases-untouched",
      "description": "phase-01 and phase-02 directories and their contents were not modified, moved, or deleted",
      "max_score": 8
    },
    {
      "name": "phase-readme-has-gate",
      "description": "The phase-03 README.md includes a gate section with a concrete pass/fail shell command (not vague language)",
      "max_score": 10
    },
    {
      "name": "task-count-matches-scope",
      "description": "The number of task files in phase-03/tasks/ is appropriate for the 6 endpoints described — tasks are not over-bundled (one task per unrelated endpoint or file is ideal)",
      "max_score": 7
    },
    {
      "name": "task-ids-continue-sequence",
      "description": "Task identifiers use P03 prefix (e.g. task-P03T01-*.md) with zero-padded 1-based numbering",
      "max_score": 7
    },
    {
      "name": "task-scoped-to-single-handler",
      "description": "Each task file covers a single route handler file (or a closely related pair) and does not bundle unrelated endpoints into one task without justification",
      "max_score": 8
    },
    {
      "name": "task-verification-is-runnable",
      "description": "Every task file includes a verification section with a specific shell command (e.g. curl -s localhost:3000/posts | jq length, or npm test -- --grep) that exits 0 on success",
      "max_score": 10
    },
    {
      "name": "no-vague-verification",
      "description": "No task verification uses phrases like 'the endpoint works', 'tests pass', or 'API responds correctly' without specifying the exact command and expected output",
      "max_score": 8
    },
    {
      "name": "dependencies-declared",
      "description": "The phase-03 README.md explicitly declares dependencies on phase-02 (data model) and any specific artefacts needed",
      "max_score": 7
    },
    {
      "name": "scaffold-scripts-used",
      "description": "Agent used new-phase.sh and new-task.sh scripts to create the phase and task files rather than creating them manually",
      "max_score": 8
    },
    {
      "name": "validate-plan-run",
      "description": "Agent ran validate-plan.sh after adding phase-03 and fixed any issues before completing",
      "max_score": 8
    },
    {
      "name": "root-readme-updated",
      "description": "The plan root README.md was updated to include phase-03 in its phase listing (additive — prior phases still listed)",
      "max_score": 7
    },
    {
      "name": "slug-format-correct",
      "description": "All new directory names and file slugs use lowercase kebab-case",
      "max_score": 4
    },
    {
      "name": "completion-summary-reported",
      "description": "Agent reported a completion summary listing the new phase README and all new task files",
      "max_score": 2
    }
  ]
}

evals

SKILL.md

tile.json