Tools, resources, and prompt

Sixteen tools, two resources, one prompt. Three tools score user stories for Integrated Vindex and thirteen belong to Agent Reflection. The server name is vindex. Every string below is what the assistant sees; schemas are published as tool input and output schemas so clients can validate calls.

Server instructions

The server advertises these instructions to every client. They are why an assistant scores several stories one at a time and leads with the finding.

instructions

Vindex scores user stories against INVEST and returns percentages with reasoning.
Score one story per call. If the user has several stories, call score_story once
for each, in sequence, and present each result as it arrives. Pass the story's
tracker id so re-scores stay free. Present the finding first, then the percentages.

`score_story`

Scores one user story synchronously and returns the INVEST result. One POST /api/v1/score to the Vindex gateway per call; the score comes back in the same call.

readOnlyHint: false
destructiveHint: false
idempotentHint: true
openWorldHint: true

Description shown to the assistant

Score a user story against the six INVEST dimensions with Vindex. Returns an overall percentage, a health band, and per-dimension percentages with reasoning and improvement suggestions. Pass story_id (the Jira key, Azure DevOps work item id, or any stable id from the user’s tracker) so the story is counted once and every later score of it is free. Without story_id, the story is identified by its text: scoring the same text again is free, scoring edited text counts as a new story. Takes 2 to 10 seconds.

Input

json schema

{
  "type": "object",
  "properties": {
    "title":               { "type": "string", "minLength": 3,  "maxLength": 300 },
    "description":         { "type": "string", "minLength": 10, "maxLength": 20000 },
    "acceptance_criteria": { "type": "string", "maxLength": 20000 },
    "story_id":            { "type": "string", "maxLength": 200,
                             "description": "Stable id from the user's tracker. Strongly recommended." }
  },
  "required": ["title", "description"],
  "additionalProperties": false
}

Output (structured content)

Real result: story NW-108 from the Northwind sample backlog, 2026-09-17 (dimension text abbreviated)

json

{
  "story_id": "NW-108",
  "overall_percent": 38,
  "health": "at_risk",
  "dimensions": {
    "independent": { "percent": 40, "reasoning": "…", "suggestions": ["…"] },
    "negotiable":  { "percent": 60, "reasoning": "…", "suggestions": ["…"] },
    "valuable":    { "percent": 60, "reasoning": "…", "suggestions": ["…"] },
    "estimable":   { "percent": 20, "reasoning": "…", "suggestions": ["…"] },
    "small":       { "percent": 20, "reasoning": "…", "suggestions": ["…"] },
    "testable":    { "percent": 30, "reasoning": "…", "suggestions": ["…"] }
  },
  "recommendations": [
    "Add measurable acceptance criteria with baseline and target metrics (e.g., p95 scan-to-confirm time, scans/hour, error rate) and specify devices/network conditions for measurement.",
    "…"
  ],
  "identity": { "mode": "story_id" },
  "vindex": { "job_id": "f58ed328-…", "processing_ms": 22867 }
}

health is healthy (80 to 100), needs_review (60 to 79), or at_risk (0 to 59). A suggestion, not a gate.
recommendations are ordered, highest impact first. The first is the finding the assistant leads with.
identity.mode is story_id when the caller’s id was used and content_hash when the story was identified by its text.
The tool also returns a markdown text rendering (story id and percentage, band, finding, top recommendations, dimension table) for clients that do not render structured content.

Errors

Returned as tool results with isError: trueand one sentence: “Vindex did not respond in time. Try again in a moment.” (65-second ceiling), “Vindex is busy for your company, try again shortly.” (concurrency), or an allowance message when the company’s new stories are paused for the month. Re-scores of counted stories still succeed when the allowance is exhausted.

One story per call, by design

There is no batch tool. A batch stacks several inference calls behind one response and fails on client timeouts exactly when a backlog is large, and an assistant has no good way to display ten full results. If the user asks for several stories, the assistant calls score_story once per story, in sequence, and presents each result as it arrives.

`get_usage`

Reports the company’s story allowance for the current month. Reads the same records the Vindex Portal usage page shows, so the two never disagree.

readOnlyHint: true
idempotentHint: true

Input

json schema

{ "type": "object", "properties": {}, "additionalProperties": false }

Output

json

{
  "company": "Acme",
  "company_name_is_placeholder": false,
  "plan": "free",                      // free | starter | standard | pro | enterprise
  "period": { "start": "2026-09-01", "end": "2026-09-30", "timezone": "UTC" },
  "stories": { "counted_this_month": 7, "allowance": 10, "remaining": 3 },
  "state": "ok",                       // ok | exhausted
  "upgrade_url": "https://www.vindexgateway.com/pricing"
}

Text rendering: “Acme has scored 7 of Up to 10 stories this month on the free plan. 3 remain. Re-scoring counted stories is free.” When the company still carries its placeholder name the text adds a reminder to rename it in the Vindex Portal.

`explain_invest`

Returns the INVEST dimension definitions and the health bands. Served from the server itself; makes no Vindex call and uses no allowance, so the assistant can explain a result without spending anything.

readOnlyHint: true
idempotentHint: true
openWorldHint: false

Input

json schema

{
  "type": "object",
  "properties": {
    "dimension": { "type": "string",
                   "enum": ["independent", "negotiable", "valuable", "estimable", "small", "testable"] }
  },
  "additionalProperties": false
}

Without dimensionit returns all six plus the bands. With one, it returns that dimension’s definition, what Vindex looks for, and common failure patterns.

Agent Reflection tools

Thirteen tools in six features: Account, Rules, Rule sets, Governance, Validation, and Reporting. They are separate from the three story tools above. Access is set per feature, by role. Who can call each tool is below, and authentication has the table. The detailed pages are Agent Reflection.

`get_account`

Account · Any member

Who the agent works for: the company, your role, whether your company only allows approved items, and which features are open, including Track reporting. Free.

Inputs

None.

Returns

Company, plan, role (Company admin or Company user), the approved-only policy, and whether reporting is open.

`get_credit_balance`

Account · Any member

The company's shared critique credit balance and its recent critique use. Every member sees the same balance. Free.

Inputs

None.

Returns

The balance and recent use.

`list_rules`

Rules · Any member

Lists the rules you can use, or reads one. Free. Returns active rules from your own catalog, your approved list, and the public catalog, and nothing else.

Inputs

query, tags: search and filter.
source: own, approved, or public.
status: for your own rules only.
rule_id, version_id: read one rule and its content.
Paging.

Returns

For each rule: rule_id, name, description, tags, status, owner, catalog, active_version_id, and modified_at.
Content is returned only when you read one rule.

Rule format

`create_rule`

Rules · Any member

Creates a rule in your private catalog as a draft. The agent should show the person the Markdown and get agreement before uploading, unless it is acting on its own.

Inputs

name, description, tags, content (the Markdown as text).

Returns

rule_id, version_id, and status: draft, with the next step.

Rule format

`update_rule`

Rules · Members of the owning company

Saves a new version of a rule, or changes its status or metadata. A new version does not become active until someone sets the rule to active.

Inputs

rule_id.
content: saves a new version.
base_version_id: the version you started from. If a newer one exists, your change is still saved and the result warns you.
status: draft, test, active, disabled, or archived. Metadata: name, description, tags.

Returns

The new version, and the company's rule sets that still pin an older version.

Rule format

`list_rulesets`

Rule sets · Any member

Lists the rule sets you can use, or reads one with its pins. Free. Reading one shows, for each pin, whether the rule has a newer active version.

Inputs

query, tags, source, paging, or a rule set ID to read.

Returns

Each rule set with its status, owner, and active version. Reading one returns its pins with each rule's name, description, owner, and version.

Rule sets

`create_ruleset`

Rule sets · Any member

Creates a rule set in your catalog as a draft. A pin with no version uses the rule's active version, and the result says which version it used.

Inputs

name, description, tags, pins (each a rule_id with an optional version_id).

Returns

ruleset_id, version_id, status: draft, and the pins as saved.

Rule sets

`update_ruleset`

Rule sets · Members of the owning company

Saves a new rule set version with changed pins, or changes status, metadata, or the active version without a new version. Archives a set.

Inputs

ruleset_id.
pins: saves the next version, in draft.
status, metadata, active version, replaced_by.

Returns

The saved version and its status.

Rule sets

`approve_item`

Governance · Company admins

Adds or removes an item on your company's approved list, pinned to a version. When your company turns on approved-only, its agents can run only approved items. Others are refused as not approved by your company.

Inputs

The rule or rule set, and the version.

Returns

The updated approved list entry.

`validate`

Validation · Any member with credits

Checks an output against one rule or one rule set and returns a pass or fail for each rule, and names any rule that did not pass. On a fail, the agent reads that rule, fixes the output against it, and calls again. $0.03 per call, whether it checks one rule or a whole rule set. Each attempt is charged.

Inputs

output, and exactly one of rule_id or ruleset_id.
Optional: version_id, session_id, max_attempts, request_id, agent_id, tags, agent_model, agent_label, output_ref.

Returns

result, the versions that ran, attempt and attempts_left, credits_left, and validation_id. For a rule set, also failed_rules and rules_run.

Critique results

`get_reporting_schema`

Reporting · Companies with Track

The metrics, dimensions, filters, and limits you may use, with definitions.

Inputs

None.

Returns

The schema and its version.

Reporting

`query_report`

Reporting · Companies with Track

Rows for one metric, scoped to your company and your role, with the metric's definition and the time the data runs to.

Inputs

metric, up to two dimensions, filters, from and to, period, compare, sort, limit, cursor.

Returns

Rows, counts, the definition, and data_as_of.

Reporting

`get_benchmarks`

Reporting · Companies with Track

Your figure on a public rule or rule set against anonymised peers, only when enough other companies use it. It never shows another company's figure.

Inputs

metric, one public rule_id or ruleset_id, a month, and optionally by_agent_model.

Returns

Your figure, the peer median, and your quartile, or a refusal that says why there is no comparison.

Reporting

For the critique loop, the instruction to give your agent is to read each rule that failed, revise the output against it, send the whole revised output in the same session, never resend unchanged output, and stop at a pass, at the attempt cap, or when credits run out, then tell the person what happened. The standing instruction on Connect your agent also asks the agent to send agent_model on every call.

Resources

URI	MIME	Content
vindex://invest/dimensions	application/json	The same content as explain_invest with no argument, for clients that load reference material as a resource
vindex://health-bands	application/json	{ healthy: [80,100], needs_review: [60,79], at_risk: [0,59] } with labels

Resources are static in content, but the endpoint still requires a valid token like every request.

Prompt: `score-story`

Argument: story_id(optional). Clients surface prompts as slash commands or quick actions; this is the one-click “score whatever is in the chat” action.

prompt text

Score the user story currently in this conversation with Vindex.

Take the story exactly as it stands right now, whether the user pasted it, you
drafted it, or you wrote it together. Do not change a word. Split it into:
- title: the story's title or one-line summary
- description: the body of the story
- acceptance_criteria: the acceptance criteria, if there are any

Call the score_story tool with those fields, passing story_id when the user has
a tracker id for the story.

Present the result in this order:
1. The finding in plain language, first.
2. The overall percentage and the health band (Healthy, Needs review, At risk).
3. The per-dimension detail only if the user asks for it.

Vindex judges the story as written. Do not offer to rewrite or improve it unless
the user asks.

The prompt contains no instruction to rewrite, improve, or re-score. What the user does with the result is between the user and their assistant.

Vocabulary in every tool string

The unit is the story. Never “tickets”, never “scores” as a thing being used up.
Allowances are phrased “Up to N stories”. No per-story price appears anywhere.
Percentages only. Never a score out of ten.
Health bands: Healthy, Needs review, At risk. They are suggestions.

Server instructions

score_story

Description shown to the assistant

Input

Output (structured content)

Errors

One story per call, by design

get_usage

Input

Output

explain_invest

Input

Agent Reflection tools

get_account

get_credit_balance

list_rules

create_rule

update_rule

list_rulesets

create_ruleset

update_ruleset

approve_item

validate

get_reporting_schema

query_report

get_benchmarks

Resources

Prompt: score-story

Vocabulary in every tool string

`score_story`

`get_usage`

`explain_invest`

`get_account`

`get_credit_balance`

`list_rules`

`create_rule`

`update_rule`

`list_rulesets`

`create_ruleset`

`update_ruleset`

`approve_item`

`validate`

`get_reporting_schema`

`query_report`

`get_benchmarks`

Prompt: `score-story`