iterative-planner

State-machine driven iterative planning and execution for complex coding tasks. Cycle: Explore → Plan → Execute → Reflect → Pivot. Filesystem as persistent memory. Use for multi-file tasks, migrations, refactoring, failed tasks, or anything non-trivial.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./src/SKILL.md

Quality

Content

73%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.

A well-engineered protocol specification with excellent workflow gating, validation feedback loops, and clean one-level reference structure. Its main weakness is conciseness: the body is long and includes verbose edge-case prose that competes with the context window.

Suggestions

Trim the Orchestrator Role Assumption and known-gap/leash-override disclosures into shorter pointers to a reference file; the core idempotency rule can be stated in 2-3 lines.

Externalize the File Lifecycle Matrix and Consolidated File Management detail into references/file-formats.md, keeping only the read-before-write rule and caps inline.

Condense the Recovery from Context Loss 13-step list into a short pointer plus the pointer-recreation command, moving the full resume procedure into a reference.

Dimension	Reasoning	Score
Conciseness	The ~494-line body is information-dense and avoids explaining concepts Claude already knows, but it carries verbose edge-case disclosures (Orchestrator Role Assumption, known gaps, leash-override exception, lengthy parentheticals) that could be tightened without losing clarity.	3 / 5
Actionability	Concrete copy-paste commands (bootstrap.mjs, validate-plan.mjs --pre-step, emit-state.mjs) and exact git commit-tag derivation rules are present, but most per-state operational detail is deferred to scripts rather than inlined, leaving minor gaps.	4 / 5
Workflow Clarity	A fully sequenced state machine with explicit transitions, validation gates (GATE:FAIL slugs), Pre/Post-Step checklists, and the Autonomy Leash revert-first feedback loop matches the anchor-5 pattern of explicit validation steps and error-recovery loops.	5 / 5
Progressive Disclosure	One-level-deep references (references/.md, scripts/modules/state-.md, agents/ip-orchestrator.md) are clearly signaled and content is appropriately split, but the SKILL.md body itself is long with inlined matrices and recovery detail that could be externalized.	4 / 5
	Total	16 / 20 Passed

Description

83%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.

A strong, third-person description that clearly states both capability and trigger conditions with concrete use-case keywords. It is held back from full marks only by slightly abstract core actions and the broad 'anything non-trivial' trigger.

Dimension	Reasoning	Score
Specificity	Names the domain ('iterative planning and execution for complex coding tasks') and concrete cycle phases (Explore → Plan → Execute → Reflect → Pivot) plus several task scenarios, but the core 'what' actions stay somewhat abstract rather than enumerating multiple concrete operations like the anchor-5 example.	4 / 5
Completeness	Explicitly answers both 'what' (state-machine iterative planning/execution with a named cycle and filesystem memory) and 'when' via a concrete 'Use for ...' trigger clause, matching the anchor-5 pattern.	5 / 5
Trigger Term Quality	'multi-file tasks, migrations, refactoring, failed tasks, or anything non-trivial' are natural phrases users say, giving good coverage, but synonyms and variations are limited and no file-extension-style triggers apply.	4 / 5
Distinctiveness Conflict Risk	The state-machine + cycle framing is a clear niche with distinct triggers, but 'anything non-trivial' is broad and creates minor overlap risk with general coding skills rather than the minimal risk of anchor 5.	4 / 5
	Total	17 / 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
referenced_paths_exist	Referenced path issues: 1 missing, 7 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: NikolasMarkou/iterative-planner
Path: src/SKILL.md
Commit: d95307d

Reviewed: 1 day 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.