feat: resolve worker reasoning effort from per-type SSM params - #99
Merged
Conversation
Add resolveClaudeEffort(workerType), mirroring resolveClaudeModel: read
/auto-graph/worker-effort/{workerType} from SSM, then the CLAUDE_EFFORT
env var, then fall through to undefined (the CLI applies its own default,
xhigh on Opus). The value is validated against low|medium|high|xhigh|max
and normalized; an invalid value warns and is ignored rather than failing
the run. runClaude appends --effort only when a level is resolved, so
omitting it preserves current behavior.
This lets operators tune reasoning effort per worker type via
aws ssm put-parameter with no redeploy — e.g. xhigh for the writer and
auditor, medium for planner/continuity-checker — to balance output
quality against token burn and rate limits. Each task role gains
ssm:GetParameter on its own /auto-graph/worker-effort/{name} param ARN.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Adds
resolveClaudeEffort(workerType), mirroring the existingresolveClaudeModel:/auto-graph/worker-effort/{workerType}from SSM →CLAUDE_EFFORTenv →undefined.low|medium|high|xhigh|maxand normalizes case/whitespace; an invalid value warns and is ignored rather than failing the run.runClaudeappends--effort <level>only when a level resolves, so omitting it preserves current behavior (the CLI's own default —xhighon Opus).ssm:GetParameteron its own/auto-graph/worker-effort/{name}ARN.Lets operators tune reasoning effort per worker via
aws ssm put-parameterwith no redeploy — balancing output quality against token burn and rate limits.Live SSM values already set
xhighhighmedium(Models remain on the bare
opusalias across all workers.)Tests
resolveClaudeEffort(SSM hit, env fallback, invalid-value rejection, no-source → undefined) and forrunClaudeappending/omitting--effort.🤖 Generated with Claude Code