Skip to content

docs(readme): add Evidence Export + Policy Simulation rows to Eval table#144

Merged
saurabhjain1592 merged 2 commits into
mainfrom
chore/readme-7row-eval-table
Apr 27, 2026
Merged

docs(readme): add Evidence Export + Policy Simulation rows to Eval table#144
saurabhjain1592 merged 2 commits into
mainfrom
chore/readme-7row-eval-table

Conversation

@saurabhjain1592

Copy link
Copy Markdown
Member

Summary

Brings the SDK README Eval Tier table from 5 rows to 7 rows, matching the four plugin READMEs.

Why: review feedback flagged that SDK READMEs showed a more compact 5-row table while plugin READMEs showed 7 rows. Audience-tuned rows are fine, but the level of detail should match across both README families.

Rows added:

  • Evidence export (CSV / JSON) — | 5,000 records · 14d window · 3/day | Unlimited
  • Policy simulation — | 300 / day | Unlimited

Why these specific rows for SDKs: both are tier-gated runtime features verified directly against platform/agent/license/tier_support.go (MaxEvidenceExportRecords, MaxEvidenceWindowDays, MaxEvidenceExportsPerDay, MaxSimulationsPerDay), and both are meaningful to SDK users moving from Community to Eval (audit/compliance teams care about Evidence Export; SDK developers testing rules care about Policy Simulation).

Plugin READMEs unchanged — already at 7 audience-tuned rows. Different rows (HITL approval gates, session overrides) for that audience.

Test plan

  • Preview renders 7-row table
  • Numbers match runtime config

Brings SDK README Eval Tier table to 7 rows, matching the 4 plugin
READMEs and the 'consistent level of detail' standard. Audience-tuned
rows differ between SDK (Concurrent executions, Pending execution
approvals — MAP/WCP-relevant for SDK readers building MAP workflows)
and plugin (HITL approval gates, session overrides — plugin-relevant
for governance hooks), but the row count is now equal across both
README families. Numbers verified against the runtime tier config.
…m capabilities

Reviewer feedback on PR: in a language-SDK README the rows for Evidence
Export and Policy Simulation could read as if they're SDK methods (e.g.,
client.exportEvidence(), client.simulatePolicy()). They are not — both
are licensed platform capabilities accessed via the AxonFlow API or the
customer portal, not via language-specific SDK helpers.

Added a Note callout below the existing footnote line that says exactly
that. Rows stay (they are real Evaluation-tier upgrade triggers worth
surfacing), the framing is now honest about how they're consumed.
@saurabhjain1592 saurabhjain1592 merged commit 07a2f26 into main Apr 27, 2026
10 checks passed
@saurabhjain1592 saurabhjain1592 deleted the chore/readme-7row-eval-table branch April 27, 2026 09:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant