-
-
Notifications
You must be signed in to change notification settings - Fork 1.4k
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
MCP(command, args=[...]) passes the whole command string as the executable — 0 tools discovered, failure silently swallowed
bugSomething isn't workingSomething isn't workingclaudeAuto-trigger Claude analysisAuto-trigger Claude analysisdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3940 In MervinPraison/PraisonAI;Agent-level
guardrails=<callable>always returns None — two conflictingTaskOutputclassesbugSomething isn't workingSomething isn't workingclaudeAuto-trigger Claude analysisAuto-trigger Claude analysisdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3939 In MervinPraison/PraisonAI;OpenAI Responses API is sent Chat Completions content parts (
type: "text"): live vision examples 400, fall back, then fail withinvalid_image_urlbugSomething isn't workingSomething isn't workingStatus: Open.#3934 In MervinPraison/PraisonAI;Public token collector stays at zero after a successful live
Agent.start(): usage tracking is gated onmetrics=True, which the constructor consolidation now defaults offbugSomething isn't workingSomething isn't workingStatus: Open.#3933 In MervinPraison/PraisonAI;Harbor Terminal-Bench adapter scores 0.000: default
max_tool_calls_per_turn=10aborts the coding loop, then a"passed"substring heuristic accepts a dummy pytestbugSomething isn't workingSomething isn't workingStatus: Open.#3932 In MervinPraison/PraisonAI;ACP IDE server constructs a tool-less chat
Agent: MCP configs, write/shell flags, and session capabilities never reach the model loopbugSomething isn't workingSomething isn't workingdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3931 In MervinPraison/PraisonAI;Headless
praisonai code -pcrashes before any LLM call:Agent.__init__()rejectsverbose=after constructor consolidationbugSomething isn't workingSomething isn't workingdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3930 In MervinPraison/PraisonAI;[Proposal] Memory for PraisonAI agents — 97.5% fewer tokens (ViBo)
documentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3899 In MervinPraison/PraisonAI;Scheduler has no change-detection ("monitor") primitive — proactive bots cannot "alert only when watched output changes"
claudeAuto-trigger Claude analysisAuto-trigger Claude analysisStatus: Open.#3872 In MervinPraison/PraisonAI;Safety gates in permissions/approval/hooks look enforced but silently no-op on real inputs
claudeAuto-trigger Claude analysisAuto-trigger Claude analysisdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3866 In MervinPraison/PraisonAI;Gateway has no commitment/promise follow-through: agent promises and user open-loops are never captured or proactively completed
bugSomething isn't workingSomething isn't workingclaudeAuto-trigger Claude analysisAuto-trigger Claude analysisStatus: Open.#3819 In MervinPraison/PraisonAI;ExecutionConfig.max_rpmis a dead field — docs and examples claim RPM limiting works, but Agent never builds aRateLimiterbugSomething isn't workingSomething isn't workingdocumentationImprovements or additions to documentationImprovements or additions to documentationStatus: Open.#3701 In MervinPraison/PraisonAI;