CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.hermes/skills/planning-with-files/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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 highly actionable with real, executable script commands and explicit validation checkpoints, and the core workflow (restore, create, validate, update) is easy to follow. However, it is overlong for a SKILL.md: host-adapter internals and padded heuristic sections belong in reference files, and half of its file references point to files missing from the shipped bundle, breaking navigation. Fixing the bundle layout would lift both conciseness and progressive disclosure.

Suggestions

Ship the referenced files or remove the links: templates/task_plan.md, templates/findings.md, templates/progress.md, reference.md, examples.md, and plan-inventory.md are cited as core steps ('Use templates/task_plan.md as reference') but are absent from the bundle.

Move the Hermes Notes section (hook behavior, max_verify_nudges defaults, Windows path resolution, plugin install commands) into a separate host-integration reference file, keeping only a one-line pointer in SKILL.md.

Drop or compress low-value sections such as 'The Core Pattern' RAM/Disk analogy and the 5-Question Reboot Test, which restate guidance already covered by the Read vs Write matrix and Critical Rules.

DimensionReasoningScore

Conciseness

Mostly operational, but the Hermes Notes section carries dense host-adapter implementation detail (pre_llm_call/post_tool_call hook behavior, agent.max_verify_nudges defaults, Windows home-path resolution), and padded sections like 'Context Window = RAM' and the 5-Question Reboot Test could be tightened or moved to a reference file.

3 / 5

Actionability

Provides copy-paste-ready bash and PowerShell blocks with Hermes-home auto-detection, concrete validation commands ('sh "$PWF_SKILL_DIR/scripts/check-complete.sh" --validate path/to/task_plan.md'), and archive/restore flags; all referenced scripts exist in scripts/. Falls short of 5 because several guidance steps point to template and reference files that are absent from the bundle.

4 / 5

Workflow Clarity

Clear restore-first sequence with explicit validation checkpoints ('Validation must succeed before substantive work and after changing phase structure') and a 3-Strike error feedback loop. Minor gaps: the parallel root-plan vs .planning/<plan>/ paths and tool-vs-script initialization options are interleaved rather than presented as one decision tree.

4 / 5

Progressive Disclosure

Sections are well-organized and references are clearly signaled, but six referenced files (templates/task_plan.md, templates/findings.md, templates/progress.md, reference.md, examples.md, plan-inventory.md) do not exist in the bundle, and dense host-specific content (Hermes Notes, install instructions, security internals) is inlined instead of split into reference files.

3 / 5

Total

14

/

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 concrete, third-person, and answers both what and when, with distinct file-based artifacts that reduce conflict risk. Its main weaknesses are a narrow trigger set that under-represents the skill's use cases and heavy reliance on internal file/script names rather than natural user vocabulary. A significant portion of the text is spent on security-boundary detail instead of capability coverage.

Suggestions

Broaden the 'when' clause to match the body's trigger set, e.g. 'Use for multi-step tasks (3+ steps), research, building projects, or any work spanning many tool calls' instead of only 'research or work needing 5+ tool calls'.

Add natural user synonyms such as 'long-running tasks', 'track progress', and 'project planning' so the description triggers on phrasing users actually say, not just file and script names.

Trim boundary detail (nonce framing, network-path disclaimers) to one short clause and use the reclaimed space to name core capabilities: creating phased plans, logging findings, and the error-retry protocol.

DimensionReasoningScore

Specificity

Names the domain and several concrete mechanisms ('Keeps task_plan.md, findings.md, and progress.md on disk', 'lifecycle hooks inject selected project planning context', '--metadata reads same-project local agent session records and emits aggregate counts'), but core capabilities from the body (phase tracking, error protocol, plan lifecycle) are not covered.

4 / 5

Completeness

Explicitly answers both 'what' (persistent file-based planning with named files and hook injection) and 'when' ('Use for research or work needing 5+ tool calls'), but the when-clause is a single narrow criterion compared to the body's broader trigger set (multi-step tasks, building projects).

4 / 5

Trigger Term Quality

Includes some natural phrases users would say ('planning', 'multi-step', 'research', 'work needing 5+ tool calls') but relies heavily on file and script names (task_plan.md, session-catchup.py) and misses common variations like 'long-running task', 'track progress', or 'project'.

3 / 5

Distinctiveness Conflict Risk

Named planning files and script names give it a mostly distinct identity, though 'planning' and 'multi-step work' broadly overlap with generic task- and todo-management skills.

4 / 5

Total

15

/

20

Passed

Validation

87%

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

CriteriaDescriptionResult

metadata_field

'metadata' should map string keys to string values

Warning

relative_links

Relative link issues: 9 missing

Warning

Total

14

/

16

Passed

Repository

OthmanAdi/planning-with-files

Path

.hermes/skills/planning-with-files/SKILL.md

Revision

51bf62d

Reviewed

Table of Contents

DiscoveryImplementationValidation

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.

Claim skill