hindsight-architect

Expert memory architect. Understands your application, identifies where memory adds value, and produces an implementation plan with bank config, tag schema, and code.

Quality

68%

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 ./skills/hindsight-architect/SKILL.md

Quality

Content

77%Weight 40%Scale 1-3

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 and workflow-clear with concrete code and well-gated phases, but it is a large monolithic file with noticeable repetition and no file-based progressive disclosure. Tightening restated points and splitting detailed reference material into bundled files would improve the two lower dimensions.

Suggestions

Consolidate the repeated "tags are for identity scoping, not content classification" and mental-model-retrieval-strategy points into a single authoritative statement to reduce token cost.

Move the detailed retain/recall/reflect parameter tables and the full Output Format templates into reference files (e.g. references/api.md, references/plan-template.md) referenced one level deep from SKILL.md.

Keep SKILL.md as a concise overview of the methodology and the three architecture decisions, linking to the reference files for product knowledge and output scaffolding.

Dimension	Reasoning	Score
Conciseness	Most content is necessary Hindsight-specific product knowledge Claude would not know, but points are restated repeatedly (tags as identity-scoping vs content-classification appears ~5 times; mental-model retrieval strategy ~4 times), so it is mostly efficient yet could be tightened rather than fully lean.	2 / 3
Actionability	Provides an executable bash preamble and concrete, copy-paste-ready Python and Node.js code for bank creation, retain, recall, mental models, and client setup; placeholders are explicitly justified by the plan-template context, matching the fully-executable anchor.	3 / 3
Workflow Clarity	Four phases are clearly sequenced with explicit gating checkpoints ("Don't move to Phase 3 until you can make the three decisions", "When approved, move to Phase 4") plus an Implementation Checklist, satisfying the clear-sequence-with-checkpoints anchor.	3 / 3
Progressive Disclosure	Sections are well organized and detailed API docs are deferred to the hindsight-docs skill, but the file is a monolithic ~40KB document with no bundle files and large inline reference material (parameter tables, output templates) that could be split out, so it sits at the some-structure-but-could-be-better level rather than the well-signaled one-level-deep level of 3.	2 / 3
	Total	10 / 12 Passed

Description

60%Weight 40%Scale 1-3

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 conveys concrete capabilities well but is missing explicit "when to use" trigger guidance and leans on Hindsight-specific jargon, which weakens trigger-term quality and distinctiveness. Adding a "Use when…" clause with natural user phrasing would raise the two capped dimensions.

Suggestions

Add a "Use when…" clause naming natural triggers, e.g. "Use when the user wants to add memory to an application, design a memory architecture, or plan how an agent should remember and learn over time."

Include plain-language keywords users would actually say ("add memory", "agent memory", "remember users", "personalization") alongside the product terms to improve trigger-term quality.

Name the product (Hindsight) in the description to sharpen distinctiveness and reduce overlap with generic memory skills.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "Understands your application, identifies where memory adds value, and produces an implementation plan with bank config, tag schema, and code" — matching the anchor for several specific concrete actions, not the level below which only names a domain plus partial actions.	3 / 3
Completeness	Clearly states WHAT the skill does but omits any "Use when…" clause or equivalent explicit trigger guidance, so per the judging guideline completeness is capped at 2 rather than reaching the explicit-trigger level of 3.	2 / 3
Trigger Term Quality	"memory" and "application" are natural user terms, but "bank config", "tag schema", and "implementation plan" are product jargon with no common variations or natural trigger phrasing; not the level above which requires good coverage of terms users would actually say.	2 / 3
Distinctiveness Conflict Risk	"memory architect" plus "bank config" and "tag schema" give a niche, but "memory" is generic, Hindsight is not named, and it could overlap with other memory-related skills; not the clear-distinct-trigger level of 3.	2 / 3
	Total	9 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (859 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: vectorize-io/hindsight
Path: skills/hindsight-architect/SKILL.md
Commit: ed120a2

Reviewed: 7 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.