/token-log - Log Phase Token Usage

Usage

/token-log STORY-XXX phase-name [input-tokens] [output-tokens]

Arguments

STORY-XXX — story ID (e.g., STORY-001, WRKF-1021)
phase-name — standardized phase name (see table below)
input-tokens — (optional) input token count
output-tokens — (optional) output token count

If tokens are not provided, prompt the user to estimate from /cost output.

Standardized Phase Names

Phase Name	Command
`pm-generate`	`/pm-generate-story`
`pm-fix`	`/pm-fix-story`
`elaboration`	`/elab-story`
`dev-setup`	Phase 0 of `/dev-implement-story`
`dev-planning`	Phase 1 of `/dev-implement-story`
`dev-implementation`	Phase 2 of `/dev-implement-story`
`dev-verification`	Phase 3 of `/dev-implement-story`
`dev-documentation`	Phase 4 of `/dev-implement-story`
`code-review`	`/dev-code-review`
`dev-fix`	`/dev-fix-story`
`qa-verify`	`/qa-verify-story`

Task

Call kb_log_tokens MCP tool with the provided arguments:

kb_log_tokens({
  story_id: "STORY-XXX",
  phase: "phase-name",
  input_tokens: {input-tokens},
  output_tokens: {output-tokens}
})

Report success with summary

Output

After logging, report the result from kb_log_tokens:

Token logged for STORY-XXX:
  Phase: phase-name
  Input: XX,XXX | Output: XX,XXX | Total: XX,XXX

Error Handling

If kb_log_tokens returns an error:

Report the error message
Suggest checking that the KB MCP server is running

If phase-name is not in the standard list:

Allow it but warn: "Non-standard phase name: 'custom-name'"

Examples

/token-log WRKF-1021 pm-generate 15000 5000
# → Logs 15k input, 5k output for pm-generate phase

/token-log WRKF-1021 elaboration 20000 2000
# → Logs 20k input, 2k output for elaboration phase
# → Cumulative now 42,000

/token-log WRKF-1021 dev-implementation 80000 37000
# → Logs 80k input, 37k output for dev-implementation
# → Cumulative now 159,000