activeagent 1.9.2 → 1.9.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2b52709687c94c3d6c989fc36bba7826b931a6d0aa6e193b1bd4482f916d945f
-  data.tar.gz: 1801338ff01afc21837db5382aee49eb4cde1f002b322b479cf2256c283e8668
+  metadata.gz: 715323643179a060b32d624ba906586778645e818c0436d4e4efe9ea6585847e
+  data.tar.gz: 3efe81e8a8d5357baf0f46adcfa5ce2fedd96234bb8ed603ffae63bac17bec58
 SHA512:
-  metadata.gz: 4d0299374c70e7e66b5edf3b141d171e79cb2f576ec32dcb6a2b4e11e9eb5bccaabbf2a5106fb5afd05f01ece8b6dc75e9a0cd956523c39ffdc4c3b5a17a0b22
-  data.tar.gz: b60f377af9288f08cd5166ba7d71edc1060ccddf23c1d2912ed4b9d105e75ce6a6325a1f0e855e2141ded3c2ebdffbb7cd0e2f37796e662cc37ef833054d21d1
+  metadata.gz: 4a0ca1235fb0e8290ad261b61d77e38b4ee87372158384644612d54e81742f9f743968c8b8fa2c4e5cbebd57b62cdba7b0b0efbede6b3ad78176a7d7cca8dc37
+  data.tar.gz: 8fc1c553e0a008287280492dc401efccf1fdd675417985ef8acd8df3cb86ebe263a7cf6066b1740c8c387c1eb6de92ffffe428c5024b33a3ddf0464e729202a3

data/CHANGELOG.md

@@ -5,6 +5,258 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.9.4] - 2026-10-25
+
+### Added
+
+- **Host links in the dashboard's user menu** (`actionagent`).
+  `ActionAgent.user_menu_links` takes an Array of `{ label:, href: }` or a
+  Proc receiving the signed-in user and the current owner, and the header's
+  avatar menu lists them above "Sign out". A hosted platform points them at
+  its workspace, members and account settings pages, and shows its admin
+  pages to the users who may use them; a self-hosted install leaves it unset
+  and sees the menu it had. Malformed entries are dropped and a raising Proc
+  is logged, so the menu seam cannot take the dashboard down.
+
+### Fixed
+
+- **Keep the trace console usable offline** (`actionagent`). The console
+  ships its own inline styles and uses native details and ordinary links
+  instead of CDN-hosted Tailwind, Turbo and Stimulus. Automatic refresh is
+  opt-in (30 or 60 seconds) and uses the browser's refresh mechanism;
+  expanding a trace no longer depends on JavaScript or a host asset pipeline.
+- **Discover and call generated agent previews** (`activeagent`). TestUnit
+  writes previews under `test/agents/previews`, with current URLs and query
+  parameters passed to the agent. Discovery checks for newly added files,
+  and the default image interceptor accepts generation objects.
+
+- **Restore Rails browser previews** (`activeagent`). `/rails/agents` lists
+  previews and renders their prompt requests without executing a provider.
+  Namespaced actions and Markdown downloads work, unknown actions return
+  404, and all endpoints enforce Rails' local-request restriction. Preview
+  directories participate in development reloading, including configured
+  paths and the legacy `test/docs/previews` directory. The dashboard gem is
+  not needed; the routes remain disabled outside development by default.
+
+- **Generate runnable agent tests** (`activeagent`). TestUnit agent tests
+  inherit from `ActiveSupport::TestCase` instead of the nonexistent
+  `ActiveAgent::TestCase`. Plain and namespaced agents, including agents
+  without actions, can load in a Rails host without calling a provider.
+
+### Changed
+
+- **One badge for every status and message role in the dashboard**
+  (`actionagent`). The agent cards' status pill and provider chip, the run
+  and sandbox session status pills (Agent Interactions, the sandbox runner)
+  and the message role pills in conversation streams all render through the
+  shared `Badge` (soft tint, strong mono text, radius 4) with one status
+  tone map (`statusTone`) and one role map (`roleBadge`), so an `active`
+  agent, a `complete` run and an `Assistant` turn look the same wherever
+  they appear and follow the dark theme, which the cards' Tailwind pills
+  did not. Two tones join the tokens for the system and developer roles.
+
+## [1.9.3] - 2026-10-21
+
+Releases `activeagent` and `actionagent` 1.9.3, with the matching browser
+sidecar: browser upload and code conversations (the sandbox browser's live
+view takes files from the person driving it, and a Claude Code or Codex
+session is a conversation with the sandbox's browser beside the chat and a
+hand-off either way, Codex on this machine's own login), evaluation run
+reports exported as JSON, HTML and Markdown with Copy as Markdown, an agent
+builder that offers only tools that work, a model leaderboard on the
+Evaluations page, complete trace window statistics, a judge pin that runs
+on its own provider, Opper as a provider and Parallel Search in the MCP
+catalog, and streaming fixes (unindexed tool call fragments, the finish
+reason of streamed and Responses generations, Anthropic cache-token usage).
+The framework, the dashboard engine and `@activeagents/browser-sidecar`
+move to 1.9.3 together.
+
+Upgrading: migration 028 adds the conversation columns to `code_sessions`.
+A host runs `bin/rails generate action_agent:install --skip` then
+`bin/rails db:migrate`, after 1.9.1's 027. Upgrade the sidecar with the
+engine: the file chooser relay and the per-turn browser tokens are in the
+1.9.3 sidecar. On a single-tenant install the new report export
+(`GET <mount>/api/evaluation_reports/:run_id`) answers 401 until
+`ActionAgent.ingest_api_key` is set. A `judge_model` pin written
+`anthropic/…` or `openai/…` now runs on that native provider; an install
+that reached such a model through OpenRouter pins `openrouter/anthropic/…`.
+
+### Added
+
+- **Export evaluation run reports as JSON, HTML and Markdown** (`activeagent`,
+  `actionagent`). `GET <mount>/api/evaluation_reports/:run_id`, with the
+  ingest key and scoped to the tenant like the POST (and refused on a
+  single-tenant install that set no key), returns a complete run
+  as the version-1 envelope the collector takes, so a run moves to another
+  install with one POST and posting it back is an identical retry (200);
+  `?format=html` returns the run's report page and `?format=md` its
+  Markdown, and `?download=1` an attachment of any of them.
+  An imported run exports under its published `run_id`, a dashboard run
+  under `dashboard-run-<id>-<created_at>`. `ActiveAgent::Evals::Publisher.envelope`
+  builds the envelope for both the publisher and the export. The dashboard's
+  run report adds `report.json` and `report.md` (one piece with
+  `?part=verdict|fix_items|scenario`), and `?download=1` saves any format
+  as `<evaluation>-run-<n>.<ext>`; the run report view and the suite's run
+  header get *Download HTML / JSON / Markdown*. New rake tasks:
+  `action_agent:evaluations:export RUN=… [EVALUATION=…] FORMAT=json|html|md [FILE=…]`
+  and `action_agent:evaluations:import FILE=… [ACCOUNT_ID=…]`.
+
+- **Copy as Markdown** (`activeagent`, `actionagent`). `Report#to_markdown`
+  now has the HTML page's content in GitHub-flavored Markdown: the verdict's
+  rationale, What to fix from `Report#fix_items` (replacing the old
+  Recommendations list), and every scenario's answers in a `<details>` block
+  with a per-model table; table cells escape `|` and line breaks. Its pieces
+  are public — `verdict_markdown`, `fix_items_markdown(model:)`,
+  `fix_item_markdown(item)`, `scenario_markdown(key)` — and the dashboard
+  and the standalone HTML page copy each of them: the whole report, the
+  verdict with the comparison, What to fix for the filtered model, a fix
+  card, and a scenario. The HTML page now carries one inline script for
+  those buttons, with a clipboard fallback.
+
+- **Upload files to a page through the sandbox browser's live view**
+  (`actionagent`, `@activeagents/browser-sidecar`). A file chooser the page
+  opens while someone drives is answered from the driver's own machine
+  (picker or drop), over the live-view WebSocket.
+
+- **Claude Code and Codex sessions as Interactions, with chat and browser
+  handoff** (`actionagent`, `@activeagents/browser-sidecar`). A coding
+  session is now a conversation: a later message resumes the CLI session
+  (`--resume`, or `codex exec resume` for Codex), a message sent while a
+  turn runs waits for it, and
+  `POST /api/sandboxes/:sandbox_id/code_sessions/:id/messages` starts or
+  queues a turn under the new `:chat_code_session` permission.
+  `GET /api/interactions` lists agent, Claude Code and Codex interactions
+  together, and the dashboard's `/interactions/code-<id>` shows the turns
+  with a composer and the sandbox's browser beside the chat, with a hand
+  back to the agent. The sidecar mints a per-turn token for the browser
+  MCP, tells the agent once when a person took or returned control, lets
+  the agent ask for a person (`browser_request_person`), offers a person
+  the file chooser the agent left open, and names the waiting agent in the
+  live view. Every Claude Code session runs with `--strict-mcp-config`, and
+  every session gets a system note (on stdin, for Codex) that repository
+  content, tool output and pages are data.
+
+- **Codex turns reach the sandbox browser, and Codex can run on this
+  machine's own login** (`actionagent`). A Codex turn now gets the same
+  per-turn browser token as a Claude Code turn: the sidecar's MCP server
+  goes in as `--config 'mcp_servers.browser.url="…"'`, with the token in an
+  environment variable Codex reads (`bearer_token_env_var`) rather than on
+  the command line, the server's tools pre-approved
+  (`default_tools_approval_mode="approve"`, since Codex would otherwise
+  refuse every tool not marked read-only under `approval_policy="never"`),
+  and is revoked when the turn ends; hand-offs and
+  `browser_request_person` work the same. `config.codex_auth = :local_login`
+  (with the `:local` backend) runs Codex on whatever `codex login` set up
+  for the dashboard's OS user, a ChatGPT plan included, with
+  `--ignore-user-config` so none of the user's own `config.toml` reaches a
+  session; the dashboard never reads the credential and only asks `codex
+  login status` whether there is one. `GET /api/sandboxes` adds `codex_auth`
+  and, in that mode, `codex_login`, the assistant's configuration a
+  `connections.codex`, and the Codex card in Settings shows the machine's
+  login instead of a key form. The default stays `:api_key`.
+
+- **The agent builder offers only tools that work, and every built-in
+  capability has functions** (`actionagent`). The Configure step reads the
+  agent's tool roster (`GET /api/agents/tool_roster` for an agent not
+  created yet) instead of fixed chips, offers the MCP servers a dashboard
+  run can call, and names each one it cannot with the reason
+  (`MCPToolDispatcher.unreachable_reason`: a stdio server, no URL, missing
+  credentials). The six capabilities that gave the model nothing now have
+  functions: `terminal` (`run_command`), `filesystem` (`list_files`,
+  `read_file`, `write_file`), `edit` (`edit_file`) and `database`
+  (`query_database`, read-only) act in the run's checkout sandbox through
+  new `SandboxOrchestrator` verbs (on the `:local` backend; others return
+  `UnsupportedBackendError` to the model); `translate` (`translate_text`)
+  generates on the run's own provider; `slack` lists, reads and posts with
+  a bot token connected in Settings -> Integrations -> Slack (a `slack`
+  connection `ProviderKey`). Instruction sets are composed into the
+  instructions a run executes under. `run_command`, `write_file`,
+  `edit_file` and `slack_post_message` wait for approval by default: an
+  agent that gains `terminal`, `filesystem`, `edit` or `slack` — from the
+  builder, a template, the Tools tab or the agents API — has them on its
+  `approval_required_tools` unless the save names its own list, and the
+  Tools tab's Approval switch takes them off. An agent that enabled one of
+  those capabilities before this release keeps its list as it is. A host's
+  `ActionAgent.mcp_catalog` headers now reach the MCP client (they were
+  dropped).
+
+- **Start an interaction with your own agent from Interactions and Run
+  Agents** (`actionagent`). Interactions has a "New interaction" picker and
+  Run Agents a "Your agents" list; both open the agent's runner on a fresh
+  conversation (`/agents/:id/run?new=1`). The runner lists the agent's tools
+  and MCP servers, and can run against a checkout sandbox (`sandbox_id`),
+  which the checkout tools need.
+
+- **A model leaderboard on the Evaluations page** (`actionagent`). The Pass
+  rate tile listed each model's passes on a line of its own, a dozen lines
+  for seven models; it now shows the overall rate, the leading model and a
+  "compare N models" toggle, which opens a leaderboard below the tiles: one
+  row per model with its pooled passes and rate, and one column per counted
+  evaluation (`—` where the model did not run). Models rank by the lower
+  bound of the Wilson score interval (95%), so a lucky 7/8 does not outrank
+  75/173.
+
+- **Opper as an OpenAI-compatible provider** (`activeagent`). `service:
+  "Opper"` generates and embeds through Opper's compatibility endpoint
+  (`https://api.opper.ai/v3/compat`), with the key from `OPPER_API_KEY`
+  when none is configured, as the Requesty provider does (#583).
+
+- **Parallel Search in the MCP catalog** (`actionagent`). An opt-in,
+  keyless `parallel` server (`web_search`, `web_fetch`) over HTTP; queries
+  and URLs go to parallel.ai. Every MCP HTTP request now sends
+  `User-Agent: ActionAgent/<version>` unless a header names another (#575).
+
+### Fixed
+
+- **Trace totals and timeline buckets cover the whole filtered window**
+  (`actionagent`). `GET /api/traces` adds `total_count`, `returned_count`,
+  `truncated`, a `summary` (requests, errors, average latency, throughput
+  per minute) and zero-filled `buckets` computed over every trace the
+  filters match, independently of the capped detail list, and the Traces
+  view reports truncation and averages recorded request durations instead
+  of time-bucket means (part of #384).
+
+- **A pinned evaluation judge runs on its own provider** (`actionagent`).
+  A `judge_model` pin selects the model's native provider, or the one its
+  qualifier names (`openrouter/anthropic/…`), instead of the first provider
+  with credentials; the qualifier is stripped before the request, the pin
+  wins over the credential resolver's default model, and a pinned route
+  with no credentials fails the run clearly when another provider is
+  configured. Native-to-gateway alias conversion stays out of scope (#431).
+  Upgrading: a pin written `anthropic/…` or `openai/…` now selects that
+  native provider, as candidate model names already did; an install that
+  reached such a model through OpenRouter pins `openrouter/anthropic/…`.
+
+- **Streamed tool calls without an `index` are no longer split into
+  fragments** (`activeagent`). OpenAI-compatible servers (Gemini's
+  compatibility endpoint, some Ollama and OpenRouter backends) stream tool
+  calls without the `index` that ties argument fragments to their call, and
+  the Chat provider turned every fragment into its own tool call with
+  partial arguments. A fragment now continues the call with the same `id`,
+  or the last call when it has no `id` and no function name.
+
+- **A streamed generation reports why it stopped** (`activeagent`).
+  `response.finish_reason` (and `stop_reason`), and the `finish_reason` of
+  the `prompt.active_agent` event, were `nil` for every streamed Chat and
+  Anthropic generation, since a stream has no response body to read it
+  from. The Chat provider now records the chunk's `finish_reason` and the
+  Anthropic provider the `message_delta`'s `stop_reason`, and the response
+  falls back to it.
+
+- **OpenAI Responses generations report why they stopped** (`activeagent`).
+  `response.finish_reason` (and `stop_reason`) was `nil` for every
+  Responses generation, streamed or not. It is now
+  `incomplete_details.reason` when the response was cut short
+  (`max_output_tokens`, `content_filter`), else the response's `status`
+  (`completed`).
+
+- **Anthropic usage with cache token fields no longer raises** (`activeagent`).
+  `Usage.from_provider_usage` recognised Anthropic's usage only by
+  `cache_creation` or `service_tier`; a payload carrying
+  `cache_read_input_tokens` or `cache_creation_input_tokens` without them
+  fell through to `Usage.new` and raised `UnknownAttributeError`. The cache
+  token fields now mark the payload as Anthropic's (#611).
+
 ## [1.9.2] - 2026-10-20
 
 Releases `activeagent` and `actionagent` 1.9.2, with the matching browser

data/README.md

@@ -57,6 +57,9 @@ bundle add openai
 # Requesty (uses OpenAI-compatible API)
 bundle add openai
 
+# Opper (uses OpenAI-compatible API)
+bundle add openai
+
 # DeepSeek (uses OpenAI-compatible API)
 bundle add openai
 

data/lib/active_agent/base.rb

@@ -26,7 +26,7 @@ module ActiveAgent
   #
   # @example Basic agent
   #   class MyAgent < ActiveAgent::Base
-  #     generate_with :openai, model: "gpt-4"
+  #     generate_with :openai, model: "gpt-5"
   #
   #     def greet
   #       prompt instructions: "Greet the user warmly"
@@ -112,7 +112,7 @@ module ActiveAgent
     # @return [void]
     #
     # @example
-    #   generate_with :openai, model: "gpt-4", temperature: 0.7
+    #   generate_with :openai, model: "gpt-5"
     def self.generate_with(provider_reference, **agent_options)
       self.prompt_provider = provider_reference
 

data/lib/active_agent/concerns/preview.rb

@@ -68,7 +68,7 @@ module ActiveAgent
     class << self
       # Returns all agent preview classes.
       def all
-        load_previews if descendants.empty?
+        load_previews
         descendants.sort_by { |agent| agent.name.titleize }
       end
 
@@ -111,7 +111,7 @@ module ActiveAgent
 
       def load_previews
         preview_paths.each do |preview_path|
-          Dir["#{preview_path}/**/*_preview.rb"].sort.each { |file| require file }
+          Dir["#{preview_path}/**/*_preview.rb"].sort.each { |file| require_dependency file }
         end
       end
 

data/lib/active_agent/evals/design_tokens.rb

@@ -58,6 +58,13 @@ module ActiveAgent
         "--color-info" => "#3b82f6",
         "--color-info-soft" => "#dbeafe",
         "--color-info-text" => "#1e40af",
+        # The system and developer message roles: badge tones like the four above.
+        "--color-system" => "#8b5cf6",
+        "--color-system-soft" => "#ede9fe",
+        "--color-system-text" => "#6d28d9",
+        "--color-dev" => "#14b8a6",
+        "--color-dev-soft" => "#ccfbf1",
+        "--color-dev-text" => "#0f766e",
 
         # Trace span colors (observability)
         "--span-root" => "#9ca3af",
@@ -111,7 +118,11 @@ module ActiveAgent
         "--color-error-soft" => "rgba(220,38,38,0.15)",
         "--color-error-text" => "#f87171",
         "--color-info-soft" => "rgba(59,130,246,0.15)",
-        "--color-info-text" => "#93c5fd"
+        "--color-info-text" => "#93c5fd",
+        "--color-system-soft" => "rgba(139,92,246,0.15)",
+        "--color-system-text" => "#c4b5fd",
+        "--color-dev-soft" => "rgba(20,184,166,0.15)",
+        "--color-dev-text" => "#5eead4"
       }.freeze
 
       # One CSS rule declaring `tokens` as custom properties on `scope`, with

data/lib/active_agent/evals/publisher.rb

@@ -94,11 +94,7 @@ module ActiveAgent
       # report may be a Report or its saved JSON hash. Full prompts, answers and
       # tool results are included; applications should make publication opt-in.
       def call(report:, run_id:, source:, agent_name:, suite:)
-        identities = { "run_id" => run_id, "source" => source, "agent_name" => agent_name, "suite" => suite }
-        identities.each do |key, value|
-          raise ArgumentError, "#{key} must be a nonempty string" unless value.is_a?(String) && !value.strip.empty?
-        end
-        body = encode(identities.merge("version" => 1, "report" => report_hash(report)))
+        body = encode(self.class.envelope(report: report, run_id: run_id, source: source, agent_name: agent_name, suite: suite))
         raise Error, "Evaluation report exceeds the 2 MiB delivery limit; publish a smaller selection" if body.bytesize > MAX_BYTES
 
         deliver("retain the report and run_id for retry") do
@@ -143,6 +139,24 @@ module ActiveAgent
         end
       end
 
+      # The version-1 envelope a collector stores: the four identities, the
+      # version and the report's hash. What #call sends, and what a
+      # collector's export of a stored run returns, so the export can be
+      # posted again. report may be a Report or its saved JSON hash.
+      #
+      # Raises ArgumentError for an identity that is not a nonempty string,
+      # or a report that is neither.
+      def self.envelope(report:, run_id:, source:, agent_name:, suite:)
+        identities = { "run_id" => run_id, "source" => source, "agent_name" => agent_name, "suite" => suite }
+        identities.each do |key, value|
+          raise ArgumentError, "#{key} must be a nonempty string" unless value.is_a?(String) && !value.strip.empty?
+        end
+        hash = report.to_h if report.respond_to?(:to_h) && !report.nil? && !report.is_a?(Array)
+        raise ArgumentError, "report must be a Report or its saved JSON hash" unless hash.is_a?(Hash)
+
+        identities.merge("version" => 1, "report" => hash)
+      end
+
       private
 
       # Posts +body+ to the endpoint with the bearer key and the delivery timeouts.
@@ -176,13 +190,6 @@ module ActiveAgent
         raise Error.new("Evaluation delivery failed (#{e.class}); #{guidance}", retryable: true)
       end
 
-      def report_hash(report)
-        hash = report.to_h if report.respond_to?(:to_h) && !report.nil? && !report.is_a?(Array)
-        raise ArgumentError, "report must be a Report or its saved JSON hash" unless hash.is_a?(Hash)
-
-        hash
-      end
-
       # Returns the envelope as JSON. A report holding invalid UTF-8, NaN,
       # Infinity or nesting over JSON's depth limit raises a non-retryable
       # Error, without the generator's message, which can quote the report.

data/lib/active_agent/evals/report.rb

@@ -8,6 +8,7 @@ module ActiveAgent
     # Markdown, or a self-contained HTML page (ReportHtml).
     class Report
       include ReportHtml
+      include ReportMarkdown
 
       # The judge a verdict names when the framework ranked the models by
       # pass rate itself, no judge having been available to rule on them.
@@ -307,24 +308,6 @@ module ActiveAgent
         JSON.pretty_generate(to_h, *args)
       end
 
-      def to_markdown
-        scenario_count = @results.map { |result| result.scenario.key }.uniq.size
-        lines = [ "# Evaluation — #{scenario_count} scenario#{'s' unless scenario_count == 1} × #{@models.size} model#{'s' unless @models.size == 1}", "" ]
-        label = @judge_label || @judge&.label
-        lines << (label ? "Judged by `#{label}`." : "No judge; scored on rules and expectations alone.")
-        lines << ""
-        lines.concat(summary_table)
-        lines << ""
-        lines.concat(total_lines)
-        lines << "**Best model: #{winner}**" if winner
-        lines << ""
-        lines.concat(comparison_lines)
-        lines.concat(matrix_table)
-        lines.concat(detail_lines)
-        lines.concat(recommendation_lines)
-        lines.join("\n")
-      end
-
       private
 
       # Whether a model summary's cost is an estimate — any replay priced
@@ -559,7 +542,7 @@ module ActiveAgent
           latency = stats["avg_duration_ms"] ? "#{stats['avg_duration_ms']} ms" : "—"
           percentiles = stats["p50_duration_ms"] ? "#{stats['p50_duration_ms']} / #{stats['p95_duration_ms']} ms" : "—"
           cells = [
-            "`#{label}`",
+            cell("`#{label}`"),
             Format.passes(stats["passed"], stats["scenarios"], style: :markdown),
             Format.score(stats["avg_score"]),
             latency,
@@ -568,7 +551,7 @@ module ActiveAgent
             Format.money(stats["cost"], estimated: estimated_cost?(stats)),
             per_thousand_cell(stats),
             (judge_cell(stats) if judged),
-            faults.presence || "—"
+            cell(faults.presence || "—")
           ].compact
           "| #{cells.join(' | ')} |"
         end
@@ -617,13 +600,13 @@ module ActiveAgent
         lines = [ "## Comparison", "", "| Model | Pass rate | #{confidence} | Scenarios |", "|---|---|---|---|" ]
         comparison["ranking"].each do |label|
           stats = comparison["models"][label]
-          lines << "| `#{label}` | #{rate_text(stats['pass_rate'])} | #{interval_text(stats['interval'])} | #{stats['scenarios']} |"
+          lines << "| #{cell("`#{label}`")} | #{rate_text(stats['pass_rate'])} | #{interval_text(stats['interval'])} | #{stats['scenarios']} |"
         end
         lines += [ "", "| Pair | Difference | #{confidence} | Shared scenarios | Decisive |", "|---|---|---|---|---|" ]
         comparison["pairs"].each do |pair|
           a, b = pair["models"]
           difference = pair["difference"] ? format("%+.1f points", pair["difference"]) : "—"
-          lines << "| `#{a}` vs `#{b}` | #{difference} | #{interval_text(pair['interval'])} | #{pair['scenarios']} | #{pair['decisive'] ? 'yes' : 'no'} |"
+          lines << "| #{cell("`#{a}` vs `#{b}`")} | #{difference} | #{interval_text(pair['interval'])} | #{pair['scenarios']} | #{pair['decisive'] ? 'yes' : 'no'} |"
         end
         lines += [ "", comparison["conclusion"].to_s, "",
                    "_#{confidence} from #{Format.delimited(method['resamples'])} resamples of whole scenarios (seed #{method['seed']})._", "" ]
@@ -642,16 +625,16 @@ module ActiveAgent
       # cost across every model, and in brackets what the judge spent on it.
       def matrix_table
         labels = @models.map(&:label)
-        header = [ "| Scenario | #{labels.map { |label| "`#{label}`" }.join(' | ')} | Cost |", "|---|#{labels.map { '---' }.join('|')}|---|" ]
+        header = [ "| Scenario | #{labels.map { |label| cell("`#{label}`") }.join(' | ')} | Cost |", "|---|#{labels.map { '---' }.join('|')}|---|" ]
         rows = @results.group_by { |result| result.scenario.key }.map do |key, cohort|
           cells = labels.map do |label|
             result = cohort.find { |candidate| candidate.label == label }
             next "—" unless result
 
             mark = result.passed? ? "✅" : (result.errored? ? "⚠️" : "❌")
-            [ mark, (Format.score(result.score) if result.score), result.fault&.tr("_", " ") ].compact.join(" ")
+            cell([ mark, (Format.score(result.score) if result.score), result.fault&.tr("_", " ") ].compact.join(" "))
           end
-          "| `#{key}` #{cell(cohort.first.scenario.prompt.truncate(70))} | #{cells.join(' | ')} | #{scenario_cost_cell(key)} |"
+          "| #{cell("`#{key}` #{cohort.first.scenario.prompt.truncate(70)}")} | #{cells.join(' | ')} | #{scenario_cost_cell(key)} |"
         end
         header + rows
       end
@@ -664,30 +647,11 @@ module ActiveAgent
         text
       end
 
-      # A prompt may contain " | " (ScenarioParser keeps it), which would
-      # otherwise split the table cell.
+      # A prompt may contain " | " (ScenarioParser keeps it), and an answer
+      # or a label a line break, either of which would otherwise break the
+      # table row.
       def cell(text)
-        text.to_s.gsub("|") { "\\|" }
-      end
-
-      def detail_lines
-        lines = [ "", "## Answers", "" ]
-        @results.each do |result|
-          lines << "### `#{result.scenario.key}` · `#{result.label}` · #{result.status}" \
-                   "#{" · score #{Format.score(result.score)}" if result.score}#{" · #{answer_cost(result)}" if result.replay.cost}"
-          lines << ""
-          lines << "> #{result.scenario.prompt}"
-          lines << ""
-          if result.replay.tool_calls.any?
-            lines << "Tools: #{result.replay.tool_calls.map { |call| "#{call['name']}#{' ✗' if call['error']}" }.join(', ')}"
-          end
-          lines << "Fault: #{result.fault.tr('_', ' ')} — #{result.summary} #{result.recommendation}" if result.fault
-          lines << "Error: #{result.replay.error}" if result.replay.error
-          lines << ""
-          lines << (result.replay.answer.presence || "_(no answer)_").to_s.truncate(1_500)
-          lines << ""
-        end
-        lines
+        text.to_s.gsub(/\s*\r?\n\s*/, " ").gsub("|") { "\\|" }
       end
 
       # "~$0.0243 · judge ~$0.0015": what one answer cost, and what judging
@@ -698,20 +662,6 @@ module ActiveAgent
         text += " · judge #{Format.money(judge_cost, estimated: estimated_usage?(result.judge_usage))}" if judge_cost
         text
       end
-
-      # Renders after `detail_lines`, whose trailing blank line separates the two sections.
-      def recommendation_lines
-        return [] if recommendations.empty?
-
-        lines = [ "## Recommendations", "" ]
-        recommendations.each do |entry|
-          lines << "- **#{entry['fault'].tr('_', ' ')}** ×#{entry['count']} (#{entry['scenario_keys'].join(', ')}): #{entry['recommendation']}"
-          entry["suggested_tools"].each do |tool|
-            lines << "  - suggested tool `#{tool['name']}`: #{tool['description']}"
-          end
-        end
-        lines << ""
-      end
     end
   end
 end