recall-before-task

Recall relevant memories from Hindsight before starting a task. Injects past learnings, decisions, and context into the current session.

Quality

57%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Risky

Do not use without reviewing

Optimize this skill with Tessl

npx tessl skill review --optimize ./skills/recall-before-task/SKILL.md

Quality

Discovery

57%

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 identifies a clear niche around memory recall from a specific system (Hindsight), which gives it good distinctiveness. However, it lacks an explicit 'Use when...' clause and could benefit from more natural trigger terms that users would actually say. The actions described are somewhat abstract rather than concretely specific.

Suggestions

Add an explicit 'Use when...' clause, e.g., 'Use when starting a new task, resuming previous work, or when the user asks about past decisions, history, or previous context.'

Include more natural trigger terms users might say, such as 'remember', 'previous session', 'what did we decide', 'past work', 'history', 'prior context'.

Dimension	Reasoning	Score
Specificity	Names the domain (memory recall from Hindsight) and some actions (injects past learnings, decisions, context), but doesn't list multiple concrete specific actions beyond 'recall' and 'inject'. The actions are somewhat abstract.	2 / 3
Completeness	The 'what' is addressed (recall memories, inject past learnings/decisions/context), but the 'when' is only implied ('before starting a task') without an explicit 'Use when...' clause specifying trigger conditions. Per rubric guidelines, missing explicit trigger guidance caps completeness at 2.	2 / 3
Trigger Term Quality	Includes 'memories', 'Hindsight', 'past learnings', 'decisions', 'context' which are somewhat relevant, but misses natural user phrases like 'remember', 'what did we decide', 'previous session', 'history', or 'recall past work'. 'Hindsight' is a product-specific term that helps but limits natural matching.	2 / 3
Distinctiveness Conflict Risk	The reference to 'Hindsight' as a specific tool/system and the focus on memory recall and past session context creates a clear niche that is unlikely to conflict with other skills. This is a distinctive capability.	3 / 3
	Total	9 / 12 Passed

Implementation

57%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This is a reasonably well-structured skill for recalling memories before a task, with a clear workflow and good output format template. Its main weaknesses are the lack of concrete, executable examples of Hindsight tool calls (making it hard to know exact syntax) and the absence of error handling or validation steps in the workflow. The content could be tightened slightly but is not egregiously verbose.

Suggestions

Add a concrete example of an actual Hindsight recall tool call with specific parameters/syntax, e.g., `recall(query="implement JWT authentication in auth module")` so Claude knows the exact invocation pattern.

Add error handling guidance: what to do if the recall tool returns an error, times out, or returns contradictory memories.

Include a brief worked example showing a sample task description, the recall queries derived from it, and the resulting synthesized context block.

Dimension	Reasoning	Score
Conciseness	Mostly efficient but includes some unnecessary explanation. The 'Important' section's first bullet ('Do NOT skip the recall step even if the task seems simple. Past context prevents repeated mistakes.') explains rationale Claude doesn't need. The step descriptions could be tighter—e.g., step 1 'Parse the task description' is somewhat obvious. However, it's not egregiously verbose.	2 / 3
Actionability	Provides a clear multi-step process and a concrete output format template, but lacks executable specifics—there's no example of an actual Hindsight tool call (tool name, parameters, syntax), no example of what recalled memories look like, and no concrete example of a task description being parsed into recall queries. The guidance is structured but not copy-paste ready.	2 / 3
Workflow Clarity	Steps are clearly sequenced (parse → recall → synthesize → present → ask), but there are no validation checkpoints. What happens if the recall tool errors out? What if memories are contradictory? The 'no memories found' case is mentioned but there's no feedback loop for partial or ambiguous results. The workflow is linear without error recovery.	2 / 3
Progressive Disclosure	For a simple, single-purpose skill under 50 lines with no need for external references, the content is well-organized with clear sections (Steps, Important, output format template). The structure is appropriate for the skill's complexity.	3 / 3
	Total	9 / 12 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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: AndreJorgeLopes/devflow
Commit: b0b1bb6

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.