hindsight-memory

Long-term memory for the agent via Hindsight. Use to recall relevant past context before answering, retain durable facts as you learn them, and reflect over accumulated memory for the "why" behind a decision. Load whenever continuity across sessions matters — the user refers to earlier work, states a lasting preference, or asks a question that prior context could answer.

Quality

85%

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

Quality

Content

86%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 tight, well-structured memory skill that clearly delineates when to recall, retain, and reflect with concrete tool-call examples and good grounding guidance. The main opportunity is making the example calls more fully parameterized to reach fully executable coverage.

Suggestions

Make the recall/retain/reflect example calls more complete by showing the full argument set for a representative common case rather than a single-argument snippet, pushing actionability toward 5.

Add an explicit grounding/verification checkpoint after retain (e.g., confirm the fact was stored or note that nothing was retained) to add a light feedback loop without overcomplicating a non-destructive skill.

Dimension	Reasoning	Score
Conciseness	The body is lean — short intro, three focused sections each with one illustrative tool call, and a bank-scope note — with no padding or explanation of concepts Claude already knows; every section earns its place.	5 / 5
Actionability	It gives concrete, copy-ready tool invocations (recall/retain/reflect with argument examples) and clear inclusion/exclusion criteria, but the examples are illustrative snippets rather than fully parameterized calls covering the common cases, leaving a minor gap versus the top anchor.	4 / 5
Workflow Clarity	Each operation is presented as a distinct, clearly sequenced when-to-use section with grounding guidance ('say when nothing relevant was found rather than inventing continuity'), but there is no explicit validate/retry feedback loop; these are read/write memory calls rather than destructive or batch operations, so the destructive-cap does not apply.	4 / 5
Progressive Disclosure	This is a single short file with no bundle references; the four well-headed sections (recall/retain/reflect/bank scope) organize a simple skill cleanly, satisfying the under-50-lines/no-external-references exception for the top score.	5 / 5
	Total	18 / 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, trigger-rich description that clearly answers both what the skill does and when to use it across the three memory operations. It could be sharpened with more memory-specific synonyms and a fuller enumeration of retrieval capabilities to reach the top of every dimension.

Dimension	Reasoning	Score
Specificity	Names the domain (long-term memory via Hindsight) and three concrete actions — recall, retain, reflect — but does not enumerate sub-capabilities like semantic/graph/temporal retrieval, so it is above midpoint (3) but not fully comprehensive (5).	4 / 5
Completeness	It explicitly states both what it does ('recall relevant past context', 'retain durable facts', 'reflect...for the why') and when to load it ('whenever continuity across sessions matters — the user refers to earlier work, states a lasting preference, or asks a question that prior context could answer'), matching the top anchor.	5 / 5
Trigger Term Quality	Natural phrases like 'refers to earlier work', 'lasting preference', and 'a question that prior context could answer' match what users say, but it lacks explicit memory-specific synonyms ('remember', 'what did we decide', 'last time'), leaving a few common terms missing.	4 / 5
Distinctiveness Conflict Risk	The Hindsight/bank framing carves a clear niche distinct from general assistance, but 'long-term memory' is a broad concept that could mildly overlap with other context-recall or note-taking skills, so it is mostly distinct with minor overlap risk rather than minimal risk.	4 / 5
	Total	17 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: vectorize-io/hindsight
Path: hindsight-integrations/agent-plugin/skills/hindsight-memory/SKILL.md
Commit: 9c7f6c6

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.