feat(agent-skill-model-effort): model/effort on skill & agent create#24
Merged
Conversation
Implements spec agent-skill-model-effort; tasks 1-7.
Adds optional --model/--effort to `csdd skill create` and --effort to
`csdd agent create` (model already existed). Effort is validated against the
closed set {low,medium,high,xhigh,max} at the create boundary before any
filesystem write, so an invalid value exits 1 and leaves no artifact; unset
keys are omitted from frontmatter so the session config is inherited (output is
byte-identical when no flag is given). Mirrors the params in the MCP shim
(csdd_skill_create gains model+effort, csdd_agent_create gains effort) and adds
model/effort-by-complexity guidance to the shipped setup commands.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
protonspy
added a commit
that referenced
this pull request
Jun 8, 2026
Surfaces the model/effort frontmatter (added by #24) in the read-only web dashboard. The session overview now reads `model`/`effort` for agents and skills into the Artifact metadata, and ResourceView renders them in the header alongside tools. Follow-up to agent-skill-model-effort. Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Lets
csdd skill createandcsdd agent createoptionally pin amodeland aneffortin the generated frontmatter, validated at the create boundary, with unset keys omitted so the session configuration is inherited. Mirrors the params in the MCP shim and teaches the shipped setup flow to pick model/effort by task complexity.csdd skill creategains--modeland--effort;csdd agent creategains--effort(model already existed).effortis validated against the closed, case-sensitive set{low, medium, high, xhigh, max}before any filesystem write — an invalid value exits1and leaves no artifact.modelis free-form (any non-empty string).inherit), so output is byte-identical when no flag is given.csdd_skill_create(model+effort) andcsdd_agent_create(effort) forward the flags via the existingflag()helper — three-surface equivalence preserved (the TUI uses the shared*CreateOptionsstruct unchanged).csdd-setup-init/csdd-setup-updatenow instruct choosing model/effort by complexity, omitting both when neither is warranted.Spec
specs/agent-skill-model-effort/— requirements, design, tasks all human-approved (ready_for_implementation = true). Implements tasks 1–7.validateEffort+ closed setmodel/effortfrontmatter lines--model/--effort+ validation beforemkdirAll--effort+ validation before writecsdd_skill_create/csdd_agent_createVerification
csdd spec test-report agent-skill-model-effort --run --lang go✅ (exit 0, coverage 81.2%, no attentions)make check(gofmt + go vet + go test -race ./...) ✅npm test(mcp-server) ✅ (74/74)go build ./cmd/csdd+ real CLI smoke test ✅ (model+effort rendered; omit byte-identical; invalid effort → exit 1, no artifact)Adversarial
code-reviewerrun: no Blockers (APPROVE WITH NITS); nits resolved (exact frontmatter byte-comparison test, error message derived from the set, explicit exit-code assertions).Risks
Low. Purely additive — pre-existing agents/skills are untouched, and no key is rewritten when flags are absent. Not security-sensitive (local file generation, closed-set validation, no auth/secrets/network/PII).
🤖 Generated with Claude Code