hivemind-memory

Global team and org memory powered by Activeloop. ALWAYS check BOTH built-in memory AND Hivemind memory when recalling information.

Quality

58%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./harnesses/claude-code/skills/hivemind-memory/SKILL.md

Quality

Content

68%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 content is action-dense, well-structured, and respects Claude's intelligence, with strong CLI reference coverage. The main weakness is missing validation/feedback loops around destructive and batch operations, which caps workflow clarity.

Suggestions

Add validation/verification steps around destructive and batch commands — e.g., run `hivemind skillify pull --dry-run` before `--force`, confirm member list before `remove`, or verify deps before `embeddings uninstall --prune`.

Consider moving the long skillify and embeddings command references into a separate reference file (e.g. references/cli.md) and keeping SKILL.md as a tighter overview with a pointer.

De-duplicate the "do NOT quote subcommands together" note so it appears once near the top of the command sections.

Dimension	Reasoning	Score
Conciseness	The body is efficient and assumes Claude's competence, with dense command references and no explanation of concepts Claude already knows; only minor trim opportunities like the repeated "do NOT quote subcommands together" warning. It is not a 5 because a couple of chatty/guarding lines could still be tightened.	4 / 5
Actionability	It provides concrete, copy-paste-ready commands across org management, skillify, and embeddings with flags documented (e.g. `hivemind skillify pull --user <email>`, `cat file.json \| jq 'keys \| length'`). It is not a 5 because dynamic placeholders (<AUTH_CMD>, <session>, <user>) and the absence of a single fully-resolved worked example leave minor gaps.	4 / 5
Workflow Clarity	The "How to Search" section is a clear 4-step sequence with a guard against jumping to raw JSONL, and a retry limit checkpoint exists. It is capped at 3 because destructive/batch operations (unpull, remove member, push, embeddings uninstall --prune) lack validation or verify-before/after feedback loops, which the rubric requires.	3 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are empty) and the body is well-organized with clear section headers plus a directory tree. It is not a 5 because at ~100 lines the dense skillify/embeddings command references could reasonably be split into a separate reference file rather than fully inlined.	4 / 5
	Total	15 / 20 Passed

Description

48%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 establishes a clear, distinct niche (Activeloop-powered org memory) with an explicit recall trigger, but is light on concrete capability actions and natural-language synonyms. Strengthening the what-side with specific verbs and broader trigger terms would lift it.

Suggestions

Add 2-3 concrete capability verbs to the what-side (e.g., "Search, recall, and capture team and org memory" instead of just naming the system).

Expand the trigger phrase with natural synonyms users actually say, e.g. "Use when the user asks to remember, recall, or look up information across the team or org."

Keep the existing explicit when-clause but pair it with a clearer 'Use when...' framing so completeness can reach 4-5.

Dimension	Reasoning	Score
Specificity	"Global team and org memory powered by Activeloop" names the domain but lists no concrete capabilities (no store/search/recall verbs as actions), matching the anchor where the domain is named but actions are minimal or generic. It is not a 3 because it does not enumerate even 1-2 concrete actions.	2 / 5
Completeness	It has an explicit when ("ALWAYS check BOTH ... when recalling information") and a what ("Global team and org memory"), but the what is vague on concrete actions, landing at the anchor with a clear what-but-weak-when or vice-versa. It is not a 4 because the what-side is too generic, and not a 2 because an explicit trigger clause is present.	3 / 5
Trigger Term Quality	It includes relevant natural terms like "memory", "team and org memory", and "recalling information", but misses common synonyms users would say ("remember", "look up", "notes") and any file extensions. It is not a 4 because keyword coverage is incomplete rather than just a few terms short.	3 / 5
Distinctiveness Conflict Risk	Naming a specific product (Activeloop) and system (Hivemind) plus "team and org memory" carves a clear niche with mostly-distinct triggers and only minor overlap with built-in memory recall. It is not a 5 because the memory-recall use case still overlaps somewhat with Claude's built-in memory skill.	4 / 5
	Total	12 / 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: activeloopai/hivemind
Path: harnesses/claude-code/skills/hivemind-memory/SKILL.md
Commit: 551d366

Reviewed: 4 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.