CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

session-agent-builder

Turn the current interactive conversation into an owner-reviewed reusable Agentlas agent; accept JSON or JSONL only for explicit terminal and automation runs.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./claude/plugins/agentlas-core-engine-meta-agent/runtime-support/skills/session-agent-builder/SKILL.md

The canonical home for this skill is session-agent-builder in agentlas-ai/Agentlas-OS

SKILL.md
Quality
Evals
Security

Quality

Content

82%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 body is an exemplary lean instruction skill: a sequenced interactive workflow with owner-review and approval checkpoints, concrete safety constraints, and an explicit terminal command. Its only weaknesses are minor — deferred detail on the package scaffold/verify gates and references to canonical files outside the bundle.

DimensionReasoningScore

Conciseness

The body is lean and directive throughout ("It is not a chronological summary", "This optional file route must never override the interactive current-session behavior") with no explanations of concepts Claude already knows and no padding. Every token earns its place, matching the anchor-5 example of assuming Claude's competence.

5 / 5

Actionability

Concrete, executable specifics are present: AGENTLAS_AGENT_HOME / ~/.agentlas/agents defaults, the IF / THEN / BECAUSE / AVOID / INSTEAD rule format, the command `session preview|merge|ir|compile --input <session.jsonl>`, and an explicit list of report contents. It falls short of 5 only because step 5 defers to the "existing package scaffold, complete, local registration, and verify gates" without specifying how those execute.

4 / 5

Workflow Clarity

The interactive route is a clear 5-step sequence with explicit checkpoints: owner review with Build/Edit regeneration before any write (step 4), writes only from the approved report (step 5), "never overwrite an existing child", and verify gates. It is not 5 because the verify/validation gates are referenced but their steps are not spelled out, leaving minor validation gaps.

4 / 5

Progressive Disclosure

The body is short and well-sectioned (Interactive route, Generalization and safety, Optional terminal route, Output) with no bundle files to organize. It loses a point because the "canonical route" pointers to `modes/session-agent-builder.md` and `agents/40-session-agent-builder/agent.md` are clearly signaled but not part of this bundle, so navigation cannot be completed from the skill alone.

4 / 5

Total

17

/

20

Passed

Description

53%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 states a clear, niche-specific capability but reads as a scope statement rather than a trigger-optimized description: it answers "what" well while omitting any "use when" guidance, and its keyword coverage lacks the natural phrasings a user would say. It is distinctive within its niche and free of fluff.

Suggestions

Add an explicit trigger clause, e.g. "Use when the owner asks to turn this session/conversation into a reusable agent, save this chat as an agent, or automate a session export in terminal runs."

Broaden natural trigger terms with synonyms such as "session", "chat", "reusable prompt", and "agent from this conversation" so the skill matches how users actually phrase the request.

Name 1-2 more concrete actions (e.g. generating a generalized session report, packaging and registering the agent) to raise specificity from a single transformation to a fuller capability list.

DimensionReasoningScore

Specificity

The description names the domain and one concrete transformation ("Turn the current interactive conversation into an owner-reviewed reusable Agentlas agent") plus the JSON/JSONL terminal constraint, but does not enumerate multiple specific actions like report generation, generalization, or registration. It matches anchor 3 (domain and 1-2 concrete actions, not comprehensive) rather than anchor 4, which requires several listed actions.

3 / 5

Completeness

The "what" is clear (convert the current conversation into an owner-reviewed reusable agent), but there is no "Use when..." clause or equivalent trigger guidance; the second clause describes input-format constraints, not usage occasions. Per the rubric guideline, a missing explicit trigger caps completeness at 3, and it cannot be 4 since "when" is entirely unstated rather than merely implicit.

3 / 5

Trigger Term Quality

Relevant keywords are present ("conversation", "agent", "terminal", "automation", "JSON or JSONL") but common natural variations users would say are missing (e.g. "session", "chat", "turn this conversation into an agent", "reusable prompt"). This fits anchor 3 (some relevant keywords, missing variations or synonyms), not 4, because natural-phrase coverage is thin.

3 / 5

Distinctiveness Conflict Risk

"Agentlas agent" and "owner-reviewed reusable agent from the current interactive conversation" carve a distinct niche with only minor overlap risk against generic agent- or prompt-building skills. It is not 5 because the absence of explicit trigger phrases leaves some residual ambiguity against closely related skills.

4 / 5

Total

13

/

20

Passed

Validation

100%

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

agentlas-ai/Agentlas-OS

Path

claude/plugins/agentlas-core-engine-meta-agent/runtime-support/skills/session-agent-builder/SKILL.md

Reviewed

Table of Contents

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.