planning-with-files

Persistent file-based planning for multi-step AI-agent work. Keeps task_plan.md, findings.md, and progress.md on disk; lifecycle hooks inject selected project planning context. Automatic recovery reads project planning files only. Explicit session-catchup.py --metadata reads same-project local agent session records and emits aggregate counts only; --replay may emit bounded nonce-framed excerpts. Optional gated mode can request continuation only when the host supports it and never runs commands declared in Markdown. The skill has no network upload path. Use for research or work needing 5+ tool calls.

Quality

72%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.codebuddy/skills/planning-with-files/SKILL.md

Quality

Content

67%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-structured, largely actionable workflow with explicit error-recovery loops and a sensible overview-plus-references layout. Its weaknesses are redundancy and conceptual padding (Core Pattern metaphor, 5-Question Reboot Test, anti-pattern rows duplicating stated rules) plus a missing concrete example for the first, mandatory resolve-plan-dir step and broken/missing template paths.

Suggestions

Delete or move to reference.md the conceptual framing Claude already knows ('The Core Pattern' RAM/disk metaphor, the 5-Question Reboot Test), and remove Anti-Patterns rows that restate rules 1 and 6 verbatim.

Add a concrete example invocation of scripts/resolve-plan-dir.sh (and .ps1) with PLAN_ID and PWF_PLAN_ROOT, since it is the first mandatory step but currently has none.

Fix reference paths: point links at references/reference.md and references/examples.md, and either include the templates/ directory in the bundle or remove the template links.

Dimension	Reasoning	Score
Conciseness	Mostly efficient (tables for file purposes, decision matrix, anti-patterns), but 'The Core Pattern' section explains context-window volatility that Claude already knows, the 5-Question Reboot Test is a padded self-check, and the Anti-Patterns table restates rules 1 and 6 nearly verbatim ('Start executing immediately → Create plan file FIRST', 'Repeat failed actions → Track attempts, mutate approach'). More than minor over-explanation, so not 4; still substantially efficient, so not 2.	3 / 5
Actionability	Concrete, copy-paste commands for session-catchup (bash and PowerShell), 'scripts/init-session.sh "Task Name"', and 'set-active-plan.sh --list' cover the main paths. Not 5 because resolve-plan-dir.sh — the first mandatory instruction — is described but never shown with an example invocation, and template links point to a templates/ directory that does not exist in the bundle.	4 / 5
Workflow Clarity	A clear sequence exists (restore state → resolve/init plan → create files → re-read before decisions → update after act) with explicit error-recovery feedback loops (3-Strike protocol ending in escalation to the user) and a validation tool (check-complete.sh). Not 5 because the steps are scattered across five separate sections and some are directive but vague ('Create missing planning files only').	4 / 5
Progressive Disclosure	Good structure: overview body with one-level-deep references to reference.md and examples.md (both present in references/) plus a scripts section. Minor gaps keep it below 5: the reference links omit the references/ path prefix, and templates/task_plan.md, findings.md, and progress.md are linked but no templates/ directory exists in the bundle.	4 / 5
	Total	15 / 20 Passed

Description

67%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and answers both what and when with a distinct identity anchored in named planning files and scripts. Its main weakness is trigger-term quality: the when-clause ('research or work needing 5+ tool calls') uses few natural user phrasings and omits common synonyms for planning and progress tracking. Third-person voice is used consistently throughout.

Suggestions

Expand the 'Use for...' clause with natural trigger phrases users would actually say, e.g. 'Use when the user asks to plan a multi-step project, break a complex task into phases, track progress across sessions, or resume unfinished work.'

Trim implementation-detail sentences (gated-mode continuation semantics, nonce-framed excerpts) that do not help a user or Claude decide when to invoke the skill, and replace them with capability verbs like 'creates, updates, and recovers task plans'.

Add common synonyms such as 'task tracking', 'working memory', and 'todo plan' to broaden natural keyword coverage.

Dimension	Reasoning	Score
Specificity	Names several concrete actions and artifacts ('Keeps task_plan.md, findings.md, and progress.md on disk', 'lifecycle hooks inject selected project planning context', 'session-catchup.py --metadata ... emits aggregate counts only'). Not 5 because coverage is implementation-detail-heavy rather than a comprehensive set of user-facing capabilities.	4 / 5
Completeness	Both 'what' (file-based planning, hook injection, recovery, gated mode) and an explicit 'when' ('Use for research or work needing 5+ tool calls') are present. Not 5 because the 'when' clause is narrow and lacks natural concrete trigger phrases; not 3 because the 'when' is explicit rather than merely implied.	4 / 5
Trigger Term Quality	'planning', 'multi-step', 'research', and 'work needing 5+ tool calls' give some relevant keywords, but the natural phrases users would actually say ('plan this project', 'break this down', 'track progress', 'resume work') are missing, and '5+ tool calls' is not user vocabulary. Fits the anchor for some relevant keywords missing common variations better than anchor 4.	3 / 5
Distinctiveness Conflict Risk	Named artifacts (task_plan.md, findings.md, progress.md, session-catchup.py) establish a clear niche with distinct triggers and minimal conflict risk beyond minor overlap with generic task-management skills.	4 / 5
	Total	15 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 5 missing	Warning

	Total	14 / 16 Passed

Repository: OthmanAdi/planning-with-files
Path: .codebuddy/skills/planning-with-files/SKILL.md
Commit: dab9d16

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.